Deepak Pathak

EDBT 2026 Demo / reviewers in the wild / expert

dblp:155/9860 · DBLP profile ↗

← Back

72ranked-venue papers

9as first author

46since 2021 · last 2025

—ORCID · conflict

Domains — the database's venue-derived domains; a paper can count in several

Artificial intelligence and machine learning · 63 · 7 first-author · 40 since 2021Graphics, computer vision, multimedia, augmented reality and games · 16 · 4 first-author · 11 since 2021Applied, interdisciplinary, general and emerging computing · 11 · 1 first-author · 5 since 2021Systems, architecture and hardware · 8 · 8 since 2021Human-computer interaction and ubiquitous computing · 2

Year	Publication	Venue	Position
2025	Generative Classifiers Avoid Shortcut Solutions abstract Discriminative approaches to classification often learn shortcuts that hold in-distribution but fail even under minor distribution shift. This failure mode stems from an overreliance on features that are spuriously correlated with the label. We show that generative classifiers, which use class-conditional generative models, can avoid this issue by modeling all features, both core and spurious, instead of mainly spurious ones. These generative classifiers are simple to train, avoiding the need for specialized augmentations, strong regularization, extra hyperparameters, or knowledge of the specific spurious correlations to avoid. We find that diffusion-based and autoregressive generative classifiers achieve state-of-the-art performance on five standard image and text distribution shift benchmarks and reduce the impact of spurious correlations in realistic applications, such as medical or satellite datasets. Finally, we carefully analyze a Gaussian toy setting to understand the inductive biases of generative classifiers, as well as the data properties that determine when generative classifiers outperform discriminative ones. Alexander C. Li, Ananya Kumar, Deepak Pathak	ICLR	3
2025	Local Policies Enable Zero-Shot Long-Horizon Manipulation abstract Sim2real for robotic manipulation is difficult due to the challenges of simulating complex contacts and generating realistic task distributions. To tackle the latter problem, we introduce ManipGen, which leverages a new class of policies for sim2real transfer: local policies. Locality enables a variety of appealing properties including invariances to absolute robot and object pose, skill ordering, and global scene configuration. We combine these policies with foundation models for vision, language and motion planning and demonstrate SOTA zero-shot performance of our method to Robosuite benchmark tasks in simulation (97 %). We transfer our local policies from simulation to reality and observe they can solve unseen long-horizon manipulation tasks with up to 8 stages with significant pose, object and scene configuration variation. ManipGen outperforms SOTA approaches such as SayCan, Open VLA, LLMTrajGen and VoxPoser across 50 real-world manipulation tasks by 36%, 76%, 62% and 60% respectively. Video results at mihdalal.github.io/manipgen Murtaza Dalal, Walter Talbott, Chen Chen 0032, Deepak Pathak, Jian Zhang 0050, Ruslan Salakhutdinov	ICRA	5
2025	Neural MP: A Neural Motion Planner abstract The current paradigm for motion planning generates solutions from scratch for every new problem, which consumes significant amounts of time and computational resources. For complex, cluttered scenes, motion planning approaches can often take minutes to produce a solution, while humans are able to accurately and safely reach any goal in seconds by leveraging their prior experience. We seek to do the same by applying data-driven learning at scale to the problem of motion planning. Our approach builds a large number of complex scenes in simulation, collects expert data from a motion planner, then distills it into a reactive neural policy. We then combine this with lightweight optimization to obtain a safe path for real world deployment. We perform a thorough evaluation of our method on 64 motion planning tasks across four diverse environments with randomized poses, scenes and obstacles, in the real world, demonstrating an improvement of 23%, 17% and 79% motion planning success rate over state of the art sampling, optimization and learning based planning methods. All code, models and datasets will be released on acceptance. Video results available at mihdalal.github.io/neuralmotionplanner. Murtaza Dalal, Russell Mendonca, Youssef Khaky, Ruslan Salakhutdinov, Deepak Pathak	IROS	6
2025	Diffusion Beats Autoregressive in Data-Constrained Settings abstract Autoregressive (AR) models have long dominated the landscape of large language models, driving progress across a wide range of tasks. Recently, diffusion-based language models have emerged as a promising alternative, though their advantages over AR models remain underexplored. In this paper, we systematically study masked diffusion models in data-constrained settings—where training involves repeated passes over limited data—and find that they significantly outperform AR models when compute is abundant but data is scarce. Diffusion models make better use of repeated data, achieving lower validation loss and superior downstream performance. We find new scaling laws for diffusion models and derive a closed-form expression for the critical compute threshold at which diffusion begins to outperform AR. Finally, we explain why diffusion models excel in this regime: their randomized masking objective implicitly trains over a rich distribution of token orderings, acting as an implicit data augmentation that AR’s fixed left-to-right factorization lacks. Our results suggest that when data, not compute, is the bottleneck, diffusion models offer a compelling alternative to the standard AR paradigm. Our code is available at: https://diffusion-scaling.github.io Mihir Prabhudesai, Mengning Wu, Katerina Fragkiadaki, Deepak Pathak	NeurIPS	5
2024	Language Models as Black-Box Optimizers for Vision-Language Models abstract Vision-language models (VLMs) pre-trained on web-scale datasets have demonstrated remarkable capabilities on downstream tasks when fine-tuned with minimal data. However, many VLMs rely on proprietary data and are not open-source, which restricts the use of white-box approaches for fine-tuning. As such, we aim to develop a black-box approach to optimize VLMs through natural language prompts, thereby avoiding the need to access model parameters, feature embeddings, or even output logits. We propose employing chat-based LLMs to search for the best text prompt for VLMs. Specifically, we adopt an automatic “hill-climbing” procedure that converges to an effective prompt by evaluating the performance of current prompts and asking LLMs to refine them based on textual feedback, all within a conversational process without human-in-the-loop. In a challenging 1-shot image classification setup, our simple approach surpasses the white-box continuous prompting method (CoOp) by an average of1.5% across 11 datasets including ImageNet. Our approach also outperforms both human-engineered and LLM-generated prompts. We high-light the advantage of conversational feedback that incor-porates both positive and negative prompts, suggesting that LLMs can utilize the implicit “gradient” direction in textual feedback for a more efficient search. In addition, we find that the text prompts generated through our strategy are not only more interpretable but also transfer well across different VLM architectures in a black-box manner. Lastly, we demonstrate our framework on a state-of-the-art black-box VLM (DALL-E 3) for text-to-image optimization. Shihong Liu, Samuel Yu, Zhiqiu Lin, Deepak Pathak, Deva Ramanan	CVPR	4
2024	SPIN: Simultaneous Perception, Interaction and Navigation abstract While there has been remarkable progress recently in the fields of manipulation and locomotion, mobile manip-ulation remains a long-standing challenge. Compared to locomotion or static manipulation, a mobile system must make a diverse range of long-horizon tasks feasible in un-structured and dynamic environments. While the applications are broad and interesting, there are a plethora of chal-lenges in developing these systems such as coordination be-tween the base and arm, reliance on onboard perception for perceiving and interacting with the environment, and most importantly, simultaneously integrating all these parts to-gether. Prior works approach the problem using disentangled modular skills for mobility and manipulation that are trivially tied together. This causes several limitations such as compounding errors, delays in decision-making, and no whole-body coordination. In this work, we present a re-active mobile manipulation framework that uses an active visual system to consciously perceive and react to its en-vironment. Similar to how humans leverage whole-body and hand-eye coordination, we develop a mobile manipu-lator that exploits its ability to move and see, more specifically - to move in order to see and to see in order to move. This allows it to not only move around and interact with its environment but also, choose “when” to perceive “what” using an active visual system. We observe that such an agent learns to navigate around complex cluttered sce-narios while displaying agile whole-body coordination using only ego-vision without needing to create environment maps. Videos are available at https://spin-robot.github.io Shagun Uppal, Ananye Agarwal, Haoyu Xiong, Kenneth Shaw, Deepak Pathak	CVPR	5
2024	Evaluating Text-to-Visual Generation with Image-to-Text Generation Zhiqiu Lin, Deepak Pathak, Xide Xia, Graham Neubig, Pengchuan Zhang, Deva Ramanan	ECCV (9)	2
2024	Meta-Evolve: Continuous Robot Evolution for One-to-many Policy Transfer abstract We investigate the problem of transferring an expert policy from a source robot to multiple different robots. To solve this problem, we propose a method named Meta-Evolve that uses continuous robot evolution to efficiently transfer the policy to each target robot through a set of tree-structured evolutionary robot sequences. The robot evolution tree allows the robot evolution paths to be shared, so our approach can significantly outperform naive one-to-one policy transfer. We present a heuristic approach to determine an optimized robot evolution tree. Experiments have shown that our method is able to improve the efficiency of one-to-three transfer of manipulation policy by up to 3.2$\times$ and one-to-six transfer of agile locomotion policy by 2.4$\times$ in terms of simulation cost over the baseline of launching multiple independent one-to-one policy transfers. Supplementary videos available at the project website: https://sites.google.com/view/meta-evolve. Xingyu Liu 0001, Deepak Pathak, Ding Zhao	ICLR	2
2024	Revisiting the Role of Language Priors in Vision-Language Models abstract Vision-language models (VLMs) are impactful in part because they can be applied to a variety of visual understanding tasks in a zero-shot fashion, without any fine-tuning. We study $\textit{generative VLMs}$ that are trained for next-word generation given an image. We explore their zero-shot performance on the illustrative task of image-text retrieval across nine popular vision-language benchmarks. Our first observation is that they can be repurposed for discriminative tasks (such as image-text retrieval) by simply computing the match score of generating a particular text string given an image. We call this probabilistic score the Visual Generative Pre-Training Score (VisualGPTScore). While the VisualGPTScore produces near-perfect accuracy on some retrieval benchmarks, it yields poor accuracy on others. We analyze this behavior through a probabilistic lens, pointing out that some benchmarks inadvertently capture unnatural language distributions by creating adversarial but unlikely text captions. In fact, we demonstrate that even a "blind" language model that ignores any image evidence can sometimes outperform all prior art, reminiscent of similar challenges faced by the visual-question answering (VQA) community many years ago. We derive a probabilistic post-processing scheme that controls for the amount of linguistic bias in generative VLMs at test time without having to retrain or fine-tune the model. We show that the VisualGPTScore, when appropriately debiased, is a strong zero-shot baseline for vision-language understanding, oftentimes producing state-of-the-art accuracy. Zhiqiu Lin, Deepak Pathak, Pengchuan Zhang, Deva Ramanan	ICML	3
2024	SAPG: Split and Aggregate Policy Gradients abstract Despite extreme sample inefficiency, on-policy reinforcement learning, aka policy gradients, has become a fundamental tool in decision-making problems. With the recent advances in GPU-driven simulation, the ability to collect large amounts of data for RL training has scaled exponentially. However, we show that current RL methods, e.g. PPO, fail to ingest the benefit of parallelized environments beyond a certain point and their performance saturates. To address this, we propose a new on-policy RL algorithm that can effectively leverage large-scale environments by splitting them into chunks and fusing them back together via importance sampling. Our algorithm, termed SAPG, shows significantly higher performance across a variety of challenging environments where vanilla PPO and other strong baselines fail to achieve high performance. Webpage at https://sapg-rl.github.io/. Jayesh Singla, Ananye Agarwal, Deepak Pathak	ICML	3
2024	Extreme Parkour with Legged Robots abstract Humans can perform parkour by traversing obstacles in a highly dynamic fashion requiring precise eye-muscle coordination and movement. Getting robots to do the same task requires overcoming similar challenges. Classically, this is done by independently engineering perception, actuation, and control systems to very low tolerances. This restricts them to tightly controlled settings such as a predetermined obstacle course in labs. In contrast, humans are able to learn parkour through practice without significantly changing their underlying biology. In this paper, we take a similar approach to developing robot parkour on a small low-cost robot with imprecise actuation and a single front-facing depth camera for perception which is low-frequency, jittery, and prone to artifacts. We show how a single neural net policy operating directly from a camera image, trained in simulation with large-scale RL, can overcome imprecise sensing and actuation to output highly precise control behavior end-to-end. We show our robot can perform a high jump on obstacles 2x its height, long jump across gaps 2x its length, do a handstand and run across tilted ramps, and generalize to novel obstacle courses with different physical properties. Parkour videos at https: //extreme-parkour.github.io/. Xuxin Cheng, Ananye Agarwal, Deepak Pathak	ICRA	4
2024	Open X-Embodiment: Robotic Learning Datasets and RT-X Models : Open X-Embodiment Collaboration abstract Large, high-capacity models trained on diverse datasets have shown remarkable successes on efficiently tackling downstream applications. In domains from NLP to Computer Vision, this has led to a consolidation of pretrained models, with general pretrained backbones serving as a starting point for many applications. Can such a consolidation happen in robotics? Conventionally, robotic learning methods train a separate model for every application, every robot, and even every environment. Can we instead train "generalist" X-robot policy that can be adapted efficiently to new robots, tasks, and environments? In this paper, we provide datasets in standardized data formats and models to make it possible to explore this possibility in the context of robotic manipulation, alongside experimental results that provide an example of effective X-robot policies. We assemble a dataset from 22 different robots collected through a collaboration between 21 institutions, demonstrating 527 skills (160266 tasks). We show that a high-capacity model trained on this data, which we call RT-X, exhibits positive transfer and improves the capabilities of multiple robots by leveraging experience from other platforms. The project website is robotics-transformer-x.github.io. Abigail O'Neill, Abhiram Maddukuri, Abhishek Gupta 0004, Abhishek Padalkar, Abraham Lee, Acorn Pooley, Agrim Gupta, Ajay Mandlekar, Ajinkya Jain, Albert Tung, Alex Bewley, Alex Irpan, Alexander Khazatsky, Anant Rai, Anchit Gupta, Andrew E. Wang, Anikait Singh, Animesh Garg, Aniruddha Kembhavi, Annie Xie, Anthony Brohan, Antonin Raffin, Archit Sharma, Arefeh Yavary, Arhan Jain, Ashwin Balakrishna, Ayzaan Wahid, Ben Burgess-Limerick, Bernhard Schölkopf, Blake Wulfe, Brian Ichter, Cewu Lu, Charles Xu 0003, Charlotte Le, Chelsea Finn, Chen Wang 0053, Chenfeng Xu, Cheng Chi 0001, Chenguang Huang, Christine Chan, Christopher Agia, Chuer Pan, Chuyuan Fu, Coline Devin, Danfei Xu, Daniel Morton, Danny Drieß, Daphne Chen, Deepak Pathak, Dhruv Shah, Dieter Büchler, Dinesh Jayaraman, Dmitry Kalashnikov, Dorsa Sadigh, Edward Johns, Ethan Paul Foster, Fangchen Liu, Federico Ceola, Fei Xia 0002, Feiyu Zhao, Freek Stulp, Gaoyue Zhou, Gaurav S. Sukhatme, Gautam Salhotra, Gilbert Feng, Giulio Schiavi, Glen Berseth, Gregory Kahn, Guanzhi Wang, Hao Su 0001, Haoshu Fang, Henghui Bao, Heni Ben Amor, Henrik I. Christensen, Hiroki Furuta, Homer Walke, Hongjie Fang, Huy Ha, Igor Mordatch, Ilija Radosavovic, Isabel Leal, Jacky Liang, Jad Abou-Chakra, Jaehyung Kim 0001, Jaimyn Drake, Jan Peters 0001, Jan Schneider 0007, Jasmine Hsu, Jeannette Bohg, Jeffrey T. Bingham, Jensen Gao, Jiaheng Hu, Jiajun Wu 0001, Jiankai Sun, Jianlan Luo, Jiayuan Gu, Jie Tan 0001, Jihoon Oh, Jimmy Wu, Jingpei Lu, Jitendra Malik, João Silvério, Joey Hejna, Jonathan Booher, Jonathan Tompson, Jonathan Yang, Jordi Salvador, Joseph J. Lim, Junhyek Han, Kanishka Rao, Karl Pertsch, Karol Hausman, Keegan Go, Keerthana Gopalakrishnan, Kenneth Y. Goldberg, Kendra Byrne, Kenneth Oslund, Kento Kawaharazuka, Kevin Black, Kevin Zhang 0002, Kiana Ehsani, Kiran Lekkala, Kirsty Ellis, Krishan Rana, Krishnan Srinivasan, Kuan Fang, Kunal Pratap Singh, Kuo-Hao Zeng, Kyle Hatch, Kyle Hsu, Laurent Itti, Yunliang Chen 0001, Lerrel Pinto, Li Fei-Fei 0001, Liam Tan, Linxi Fan, Lionel Ott, Lisa Lee, Luca Weihs, Magnum Chen, Marion Lepert, Marius Memmel, Masayoshi Tomizuka, Masha Itkina, Mateo Guaman Castro, Max Spero, Maximilian Du, Michael Ahn, Michael C. Yip, Mingtong Zhang 0003, Mingyu Ding, Minho Heo, Mohan Kumar Srirama, Mohit Sharma 0001, Moo Jin Kim, Naoaki Kanazawa, Nicklas Hansen 0001, Nicolas Heess, Nikhil J. Joshi, Niko Sünderhauf, Norman Di Palo, Nur Muhammad Shafiullah, Oier Mees, Oliver Kroemer, Osbert Bastani, Pannag R. Sanketi, Patrick Tree Miller, Patrick Yin, Paul Wohlhart, Peng Xu 0010, Peter David Fagan, Peter Mitrano, Pierre Sermanet, Pieter Abbeel, Priya Sundaresan, Qiuyu Chen, Rafael Rafailov, Ria Doshi, Roberto Martin Martin, Rohan Baijal, Rosario Scalise, Rose Hendrix, Roy Lin, Runjia Qian, Russell Mendonca, Rutav Shah, Ryan Hoque, Ryan Julian, Samuel Bustamante-Gomez, Sean Kirmani, Sergey Levine, Sherry Moore, Shikhar Bahl, Shivin Dass, Shubham D. Sonawani, Shuran Song, Sichun Xu, Siddhant Haldar, Siddharth Karamcheti, Simeon Adebola, Simon Guist, Soroush Nasiriany, Stefan Schaal, Stefan Welker, Stephen Tian, Subramanian Ramamoorthy, Sudeep Dasari, Suneel Belkhale, Sungjae Park, Suraj Nair 0003, Suvir Mirchandani, Takayuki Osa, Tanmay Gupta, Tatsuya Harada, Tatsuya Matsushima, Ted Xiao, Thomas Kollar, Tianhe Yu, Tianli Ding, Todor Davchev, Tony Z. Zhao, Travis Armstrong, Trevor Darrell, Trinity Chung, Vidhi Jain, Vincent Vanhoucke, Wolfram Burgard, Xiaolong Wang 0004, Xinghao Zhu, Xinyang Geng, Liangwei Xu, Yecheng Jason Ma 0001, Yejin Kim 0003, Yevgen Chebotar, Yilin Wu 0003, Yonatan Bisk, Yoonyoung Cho, Youngwoon Lee, Yuchen Cui, Yueh-Hua Wu, Yujin Tang, Yuke Zhu, Yunchu Zhang, Yunfan Jiang 0001, Yunshuang Li, Yunzhu Li, Yusuke Iwasawa, Yutaka Matsuo, Zehan Ma, Zichen Jeff Cui, Zichen Zhang 0016, Zipeng Lin	ICRA	52
2024	Multi-Modal Fusion Methods with Local Neighborhood Information for Crop Yield Prediction at Field and Subfield Levels abstract Yield prediction at both field and subfield level poses a significant challenge, yet it holds paramount importance for decision-making and food security within the agricultural sector. Recent efforts, focused on integrating remote sensing data coupled with machine learning models, thereby creating globally scalable models for various crop types. This study underscores the effectiveness of Sentinel-2 and complementary data sources such as weather, soil, and terrain in enhancing machine learning-based yield prediction. We address the limitations of previous works and introduce a framework that incorporates local neighborhood information using convolutional neural networks and geographical coordinates. Additionally, we address the complexity of sensor fusion, showcasing both input fusion and feature fusion frameworks. We highlight that handling modalities with varying spatial and temporal resolutions requires adequate and advanced fusion mechanisms in crop yield prediction. Notably, this study reports an R2of 0.86 for soybean in Argentina using a feature fusion scheme with attention mechanism. The results are demonstrated on a large yield dataset for soybean, wheat, and rapeseed distributed across Argentina, Uruguay, and Germany. Miro Miranda, Deepak Pathak, Marlon Nuske, Andreas Dengel 0001	IGARSS	2
2024	On the Surprising Effectiveness of Attention Transfer for Vision Transformers abstract Conventional wisdom suggests that pre-training Vision Transformers (ViT) improves downstream performance by learning useful representations. Is this actually true? We investigate this question and find that the features and representations learned during pre-training are not essential. Surprisingly, using only the attention patterns from pre-training (i.e., guiding how information flows between tokens) is sufficient for models to learn high quality features from scratch and achieve comparable downstream performance. We show this by introducing a simple method called attention transfer, where only the attention patterns from a pre-trained teacher ViT are transferred to a student, either by copying or distilling the attention maps. Since attention transfer lets the student learn its own features, ensembling it with a fine-tuned teacher also further improves accuracy on ImageNet. We systematically study various aspects of our findings on the sufficiency of attention maps, including distribution shift settings where they underperform fine-tuning. We hope our exploration provides a better understanding of what pre-training accomplishes and leads to a useful alternative to the standard practice of fine-tuning. Alexander C. Li, Yuandong Tian, Beidi Chen, Deepak Pathak, Xinlei Chen	NeurIPS	4
2023	Affordances from Human Videos as a Versatile Representation for Robotics abstract Building a robot that can understand and learn to interact by watching humans has inspired several vision problems. However, despite some successful results on static datasets, it remains unclear how current models can be used on a robot directly. In this paper, we aim to bridge this gap by leveraging videos of human interactions in an environment centric manner. Utilizing internet videos of human behavior, we train a visual affordance model that estimates where and how in the scene a human is likely to interact. The structure of these behavioral affordances directly enables the robot to perform many complex tasks. We show how to seamlessly integrate our affordance model with four robot learning paradigms including offline imitation learning, exploration, goal-conditioned learning, and action parameterization for reinforcement learning. We show the efficacy of our approach, which we call Vision-Robotics Bridge (VRB) across 4 real world environments, over 10 different tasks, and 2 robotic platforms operating in the wild. Shikhar Bahl, Russell Mendonca, Unnat Jain, Deepak Pathak	CVPR	5
2023	Multimodality Helps Unimodality: Cross-Modal Few-Shot Learning with Multimodal Models abstract The ability to quickly learn a new task with minimal instruction - known as few-shot learning - is a central aspect of intelligent agents. Classical few-shot benchmarks make use of few-shot samples from a single modality, but such samples may not be sufficient to characterize an entire concept class. In contrast, humans use cross-modal information to learn new concepts efficiently. In this work, we demonstrate that one can indeed build a better visual dog classifier by reading about dogs and listening to them bark. To do so, we exploit the fact that recent multimodal foundation models such as CLIP are inherently cross-modal, mapping different modalities to the same representation space. Specifically, we propose a simple cross-modal adaptation approach that learns from few-shot examples spanning different modalities. By repurposing class names as additional one-shot training samples, we achieve SOTA results with an embarrassingly simple linear classifier for vision-language adaptation. Furthermore, we show that our approach can benefit existing methods such as prefix tuning, adapters, and classifier ensembling. Finally, to explore other modalities beyond vision and language, we construct the first (to our knowledge) audiovisual few-shot benchmark and use cross-modal training to improve the performance of both image and audio classification. Project site at link. Zhiqiu Lin, Samuel Yu, Zhiyi Kuang 0002, Deepak Pathak, Deva Ramanan	CVPR	4
2023	Your Diffusion Model is Secretly a Zero-Shot Classifier abstract The recent wave of large-scale text-to-image diffusion models has dramatically increased our text-based image generation abilities. These models can generate realistic images for a staggering variety of prompts and exhibit impressive compositional generalization abilities. Almost all use cases thus far have solely focused on sampling; however, diffusion models can also provide conditional density estimates, which are useful for tasks beyond image generation. In this paper, we show that the density estimates from large-scale text-to-image diffusion models like Stable Diffusion can be leveraged to perform zero-shot classification without any additional training. Our generative approach to classification, which we call Diffusion Classifier, attains strong results on a variety of benchmarks and outperforms alternative methods of extracting knowledge from diffusion models. Although a gap remains between generative and discriminative approaches on zero-shot recognition tasks, our diffusion-based approach has stronger multimodal compositional reasoning abilities than competing discriminative approaches. Finally, we use Diffusion Classifier to extract standard classifiers from class-conditional diffusion models trained on ImageNet. These models approach the performance of SOTA discriminative classifiers and exhibit strong "effective robustness" to distribution shift. Overall, our results are a step toward using generative over discriminative models for downstream tasks. Results and visualizations on our website: diffusion-classifier.github.io/ Alexander C. Li, Mihir Prabhudesai, Shivam Duggal, Ellis Brown, Deepak Pathak	ICCV	5
2023	Efficient RL via Disentangled Environment and Agent Representations abstract Agents that are aware of the separation between the environments and themselves can leverage this understanding to form effective representations of visual input. We propose an approach for learning such structured representations for RL algorithms, using visual knowledge of the agent, which is often inexpensive to obtain, such as its shape or mask. This is incorporated into the RL objective using a simple auxiliary loss. We show that our method, SEAR (Structured Environment-Agent Representations), outperforms state-of-the-art model-free approaches over 18 different challenging visual simulation environments spanning 5 different robots. Kevin Gmelin, Shikhar Bahl, Russell Mendonca, Deepak Pathak	ICML	4
2023	Internet Explorer: Targeted Representation Learning on the Open Web abstract Vision models typically rely on fine-tuning general-purpose models pre-trained on large, static datasets. These general-purpose models only capture the knowledge within their pre-training datasets, which are tiny, out-of-date snapshots of the Internet—where billions of images are uploaded each day. We suggest an alternate approach: rather than hoping our static datasets transfer to our desired tasks after large-scale pre-training, we propose dynamically utilizing the Internet to quickly train a small-scale model that does extremely well on a target dataset. Our approach, called Internet Explorer, explores the web in a self-supervised manner to progressively find relevant examples that improve performance on a desired target dataset. It cycles between searching for images on the Internet with text queries, self-supervised training on downloaded images, determining which images were useful, and prioritizing what to search for next. We evaluate Internet Explorer across several datasets and show that it outperforms or matches CLIP oracle performance using just a single GPU desktop to actively query the Internet for 30-40 hours. Alexander C. Li, Ellis Brown, Alexei A. Efros, Deepak Pathak	ICML	4
2023	Test-time Adaptation with Slot-Centric Models abstract Current visual detectors, though impressive within their training distribution, often fail to parse out-of-distribution scenes into their constituent entities. Recent test-time adaptation methods use auxiliary self-supervised losses to adapt the network parameters to each test example independently and have shown promising results towards generalization outside the training distribution for the task of image classification. In our work, we find evidence that these losses are insufficient for the task of scene decomposition, without also considering architectural inductive biases. Recent slot-centric generative models attempt to decompose scenes into entities in a self-supervised manner by reconstructing pixels. Drawing upon these two lines of work, we propose Slot-TTA, a semi-supervised slot-centric scene decomposition model that at test time is adapted per scene through gradient descent on reconstruction or cross-view synthesis objectives. We evaluate Slot-TTA across multiple input modalities, images or 3D point clouds, and show substantial out-of-distribution performance improvements against state-of-the-art supervised feed-forward detectors, and alternative test-time adaptation methods. Project Webpage: http://slot-tta.github.io/ Mihir Prabhudesai, Anirudh Goyal, Sujoy Paul, Sjoerd van Steenkiste, Mehdi S. M. Sajjadi, Gaurav Aggarwal, Thomas Kipf, Deepak Pathak, Katerina Fragkiadaki	ICML	8
2023	Legs as Manipulator: Pushing Quadrupedal Agility Beyond Locomotion abstract Locomotion has seen dramatic progress for walking or running across challenging terrains. However, robotic quadrupeds are still far behind their biological counterparts, such as dogs, which display a variety of agile skills and can use the legs beyond locomotion to perform several basic manipulation tasks like interacting with objects and climbing. In this paper, we take a step towards bridging this gap by training quadruped robots not only to walk but also to use the front legs to climb walls, press buttons, and perform object interaction in the real world. To handle this challenging optimization, we decouple the skill learning broadly into locomotion, which involves anything that involves movement whether via walking or climbing a wall, and manipulation, which involves using one leg to interact while balancing on the other three legs. These skills are trained in simulation using curriculum and transferred to the real world using our proposed sim2real variant that builds upon recent locomotion success. Finally, we combine these skills into a robust long-term plan by learning a behavior tree that encodes a high-level task hierarchy from one clean expert demonstration. We evaluate our method in both simulation and real-world showing successful executions of both short as well as long-range tasks and how robustness helps confront external perturbations. Videos at https://robot-skills.github.io/. Xuxin Cheng, Deepak Pathak	ICRA	3
2023	ALAN: Autonomously Exploring Robotic Agents in the Real World abstract Robotic agents that operate autonomously in the real world need to continuously explore their environment and learn from the data collected, with minimal human supervision. While it is possible to build agents that can learn in such a manner without supervision, current methods struggle to scale to the real world. Thus, we propose ALAN, an autonomously exploring robotic agent, that can perform tasks in the real world with little training and interaction time. This is enabled by measuring environment change, which reflects object movement and ignores changes in the robot position. We use this metric directly as an environment-centric signal, and also maximize the uncertainty of predicted environment change, which provides agent-centric exploration signal. We evaluate our approach on two different real-world play kitchen settings, enabling a robot to efficiently explore and discover manipulation skills, and perform tasks specified via goal images. Videos can be found at https://robo-explorer.github.io/ Russell Mendonca, Shikhar Bahl, Deepak Pathak	ICRA	3
2023	Crop Yield Prediction: An Operational Approach to Crop Yield Modeling on Field and Subfield Level with Machine Learning Models abstract Accurate and reliable crop yield prediction is a complex task. The yield of a crop depends on a variety of factors whose accurate measurement and modeling is challenging. At the same time, reliable yield prediction is highly desirable for farmers to optimize crop production. In this paper, we introduce a modeling based on remote sensing data and Machine Learning models evaluated on a large-scale dataset to address the challenge of an operational crop yield estimation and forecasting on field and subfield level. With our approach, we aim towards a global yield modeling based on Machine Learning models which operates across crop types without the need for crop-specific modeling. We demonstrate that our approach learns to map in-field variability for all studied crop types. Overall, the predictions have an error (RRMSE) of around 15% and an R2value of 0.77 at field level. Patrick Helber, Benjamin Bischke, Peter Habelitz, Cristhian Sanchez, Deepak Pathak, Miro Miranda, Hiba Najjar, Francisco Alejandro Mena, Jayanth Siddamsetty, Diego Arenas, Michaela Vollmer, Marcela Charfuelan, Marlon Nuske, Andreas Dengel 0001	IGARSS	5
2023	Feature Attribution Methods for Multivariate Time-Series Explainability in Remote Sensing abstract Numerous remote sensing applications rely on temporal satellite data, and Deep learning models are increasingly being used for such tasks. Nevertheless, these models operate as black boxes, lacking transparency and understandability. We address this gap by using explainable AI on an agricultural task. Specifically, we trained a recurrent neural network on individual pixels from multispectral time-series of Sentinel-2 satellite images to predict crop yield. We then applied nine feature attribution methods on a sample of the dataset and computed the spectral and temporal contributions to the final individual predictions. The aggregated results were evaluated qualitatively and quantitatively. Results suggest that LIME and Shapley sampling value methods performed best on the quantitative scores, followed by GradientShap. Most backpropagation-based techniques had highly inconsistent scores across the explained data points. Finally, to guide remote sensing practitioners in using Explainable AI on similar datasets, we further discuss some selection criteria to be considered. Hiba Najjar, Patrick Helber, Benjamin Bischke, Peter Habelitz, Cristhian Sanchez, Francisco Alejandro Mena, Miro Miranda, Deepak Pathak, Jayanth Siddamsetty, Diego Arenas, Michaela Vollmer, Marcela Charfuelan, Marlon Nuske, Andreas Dengel 0001	IGARSS	8
2023	Predicting Crop Yield with Machine Learning: An Extensive Analysis of Input Modalities and Models on a Field and Sub-Field Level abstract We introduce a simple yet effective early fusion method for crop yield prediction that handles multiple input modalities with different temporal and spatial resolutions. We use high-resolution crop yield maps as ground truth data to train crop and machine learning model agnostic methods at the sub-field level. We use Sentinel-2 satellite imagery as the primary modality for input data with other complementary modalities, including weather, soil, and DEM data. The proposed method uses input modalities available with global coverage, making the framework globally scalable. We explicitly highlight the importance of input modalities for crop yield prediction and emphasize that the best-performing combination of input modalities depends on region, crop, and chosen model. Deepak Pathak, Miro Miranda, Francisco Alejandro Mena, Cristhian Sanchez, Patrick Helber, Benjamin Bischke, Peter Habelitz, Hiba Najjar, Jayanth Siddamsetty, Diego Arenas, Michaela Vollmer, Marcela Charfuelan, Marlon Nuske, Andreas Dengel 0001	IGARSS	1
2023	Influence of Data Cleaning Techniques on Sub-Field Yield Predictions abstract Modern combine harvesters can collect geo-located real-time yield measurement while harvesting. This data can be used to train Machine Learning models that predict the yield at sub-field level based on remote sensing input data. The performance of these models is, however, highly dependent on the quality of the yield data. It is therefore important to develop automatic cleaning techniques to correct for common errors in combine harvester yield maps. In this work, we compare different combinations of data cleaning techniques by evaluating their impact on the yield-prediction model performance at field and sub-field level. Our findings indicate that basic cleaning techniques such as absolute thresholds are sufficient at the field level, whereas the performance at the sub-field level is enhanced through the utilization of more intricate statistical cleaning methods. Cristhian Sanchez, Deepak Pathak, Miro Miranda, Marcela Charfuelan, Patrick Helber, Marlon Nuske, Benjamin Bischke, Peter Habelitz, Nafisur Rahman, Francisco Alejandro Mena, Hiba Najjar, Jayanth Siddamsetty, Diego Arenas, Michaela Vollmer, Andreas Dengel 0001	IGARSS	2
2023	Test-time Adaptation of Discriminative Models via Diffusion Generative Feedback Mihir Prabhudesai, Tsung-Wei Ke, Alexander C. Li, Deepak Pathak, Katerina Fragkiadaki	NeurIPS	4
2023	FLAVR: Flow-Agnostic Video Representations for Fast Frame Interpolation abstract Most modern frame interpolation approaches rely on explicit bidirectional optical flows between adjacent frames, thus are sensitive to the accuracy of underlying flow estimation in handling occlusions while additionally introducing computational bottlenecks unsuitable for efficient deployment. In this work, we propose a flow-free approach that is completely end-to-end trainable for multi-frame video interpolation. Our method, FLAVR, leverages 3D spatio-temporal kernels to directly learn motion properties from unlabeled videos and greatly simplifies the process of training, testing and deploying frame interpolation models. As a result, FLAVR delivers up to 6× speed up compared to the current state-of-the-art methods for multi-frame interpolation while consistently demonstrating superior qualitative and quantitative results compared with prior methods on popular benchmarks including Vimeo-90K, Adobe-240FPS, and GoPro. Finally, we show that frame interpolation is a competitive self-supervised pre-training task for videos via demonstrating various novel applications of FLAVR including action recognition, optical flow estimation, and video object tracking. Code and trained models are provided in the supplementary material. Tarun Kalluri, Deepak Pathak, Manmohan Krishna Chandraker, Du Tran	WACV	2
2023	FLAVR: flow-free architecture for fast video frame interpolation Tarun Kalluri, Deepak Pathak, Manmohan Krishna Chandraker, Du Tran	Mach. Vis. Appl.	2
2022	Topologically-Aware Deformation Fields for Single-View 3D Reconstruction abstract We present a new framework to learn dense 3D re-construction and correspondence from a single 2D image. The shape is represented implicitly as deformation over a category-level occupancy field and learned in an unsupervised manner from an unaligned image collection without using any 3D supervision. However, image collections usually contain large intra-category topological variation, e.g. images of different chair instances, posing a major challenge. Hence, prior methods are either restricted only to categories with no topological variation for estimating shape and correspondence or focus only on learning shape independently for each instance without any correspondence. To address this issue, we propose a topologically-aware deformation field that maps 3D points in object space to a higher-dimensional canonical space. Given a single image, we first implicitly deform a 3D point in the object space to a learned category-specific canonical space using the topologically-aware field and then learn the 3D shape in the canonical space. Both the canonical shape and deformation field are trained end-to-end using differentiable rendering via learned recurrent ray marcher. Our approach, dubbed TARS, achieves state-of-the-art reconstruction fidelity on several datasets: ShapeNet, Pascal3D+, CUB, and Pix3D chairs. Shivam Duggal, Deepak Pathak	CVPR	2
2022	Coupling Vision and Proprioception for Navigation of Legged Robots abstract We exploit the complementary strengths of vision and pro-prioception to develop a point-goal navigation system for legged robots, called VP-Nav. Legged systems are capable of traversing more complex terrain than wheeled robots, but to fully utilize this capability, we need a high-level path planner in the navigation system to be aware of the walking capabilities of the low-level locomotion policy in varying environments. We achieve this by using proprioceptive feedback to ensure the safety of the planned path by sensing unexpected obstacles like glass walls, terrain properties like slipperiness or softness of the ground and robot properties like extra payload that are likely missed by vision. The navigation system uses onboard cameras to generate an occupancy map and a corresponding cost map to reach the goal. A fast marching planner then generates a target path. A velocity command generator takes this as input to generate the desired velocity for the walking policy. A safety advisor module adds sensed unexpected obstacles to the occupancy map and environment-determined speed limits to the velocity command generator. We show superior performance compared to wheeled robot baselines, and ablation studies which have disjoint high-level planning and low-level control. We also show the real-world deployment of VP-Nav on a quadruped robot with onboard sensors and computation. Videos at https://navigation-locomotion.github.io Zipeng Fu, Ashish Kumar 0007, Ananye Agarwal, Haozhi Qi, Jitendra Malik, Deepak Pathak	CVPR	6
2022	Understanding Collapse in Non-contrastive Siamese Representation Learning Alexander C. Li, Alexei A. Efros, Deepak Pathak	ECCV (31)	3
2022	Language Models as Zero-Shot Planners: Extracting Actionable Knowledge for Embodied Agents abstract Can world knowledge learned by large language models (LLMs) be used to act in interactive environments? In this paper, we investigate the possibility of grounding high-level tasks, expressed in natural language (e.g. “make breakfast”), to a chosen set of actionable steps (e.g. “open fridge”). While prior work focused on learning from explicit step-by-step examples of how to act, we surprisingly find that if pre-trained LMs are large enough and prompted appropriately, they can effectively decompose high-level tasks into mid-level plans without any further training. However, the plans produced naively by LLMs often cannot map precisely to admissible actions. We propose a procedure that conditions on existing demonstrations and semantically translates the plans to admissible actions. Our evaluation in the recent VirtualHome environment shows that the resulting method substantially improves executability over the LLM baseline. The conducted human evaluation reveals a trade-off between executability and correctness but shows a promising sign towards extracting actionable knowledge from language models. Wenlong Huang, Pieter Abbeel, Deepak Pathak, Igor Mordatch	ICML	3
2022	REvolveR: Continuous Evolutionary Models for Robot-to-robot Policy Transfer abstract A popular paradigm in robotic learning is to train a policy from scratch for every new robot. This is not only inefficient but also often impractical for complex robots. In this work, we consider the problem of transferring a policy across two different robots with significantly different parameters such as kinematics and morphology. Existing approaches that train a new policy by matching the action or state transition distribution, including imitation learning methods, fail due to optimal action and/or state distribution being mismatched in different robots. In this paper, we propose a novel method named REvolveR of using continuous evolutionary models for robotic policy transfer implemented in a physics simulator. We interpolate between the source robot and the target robot by finding a continuous evolutionary change of robot parameters. An expert policy on the source robot is transferred through training on a sequence of intermediate robots that gradually evolve into the target robot. Experiments on a physics simulator show that the proposed continuous evolutionary model can effectively transfer the policy across robots and achieve superior sample efficiency on new robots. The proposed method is especially advantageous in sparse reward settings where exploration can be significantly reduced. Xingyu Liu 0001, Deepak Pathak, Kris Makoto Kitani	ICML	2
2022	Zero-Shot Reward Specification via Grounded Natural Language abstract Reward signals in reinforcement learning are expensive to design and often require access to the true state which is not available in the real world. Common alternatives are usually demonstrations or goal images which can be labor-intensive to collect. On the other hand, text descriptions provide a general, natural, and low-effort way of communicating the desired task. However, prior works in learning text-conditioned policies still rely on rewards that are defined using either true state or labeled expert demonstrations. We use recent developments in building large-scale visuolanguage models like CLIP to devise a framework that generates the task reward signal just from goal text description and raw pixel observations which is then used to learn the task policy. We evaluate the proposed framework on control and robotic manipulation tasks. Finally, we distill the individual task policies into a single goal text conditioned policy that can generalize in a zero-shot manner to new tasks with unseen objects and unseen goal text descriptions. Parsa Mahmoudieh, Deepak Pathak, Trevor Darrell	ICML	2
2022	Adapting Rapid Motor Adaptation for Bipedal Robots abstract Recent advances in legged locomotion have en-abled quadrupeds to walk on challenging terrains. However, bipedal robots are inherently more unstable and hence it's harder to design walking controllers for them. In this work, we leverage recent advances in rapid adaptation for locomotion control, and extend them to work on bipedal robots. Similar to existing works, we start with a base policy which produces actions while taking as input an estimated extrinsics vector from an adaptation module. This extrinsics vector contains information about the environment and enables the walking controller to rapidly adapt online. However, the extrinsics estimator could be imperfect, which might lead to poor performance of the base policy which expects a perfect estimator. In this paper, we propose A-RMA (Adapting RMA), which additionally adapts the base policy for the imperfect extrinsics estimator by finetuning it using model-free RL. We demonstrate that A-RMA outperforms a number of RL-based baseline controllers and model-based controllers in simulation, and show zero-shot deployment of a single A-RMA policy to enable a bipedal robot, Cassie, to walk in a variety of different scenarios in the real world beyond what it has seen during training. Videos and results at https: //ashish-kmr.github.io/a-rma/ Ashish Kumar 0007, Zhongyu Li 0003, Jun Zeng 0002, Deepak Pathak, Koushil Sreenath, Jitendra Malik	IROS	4
2022	Continual Learning with Evolving Class Ontologies abstract Lifelong learners must recognize concept vocabularies that evolve over time. A common yet underexplored scenario is learning with class labels that continually refine/expand old classes. For example, humans learn to recognize ${\tt dog}$ before dog breeds. In practical settings, dataset ${\it versioning}$ often introduces refinement to ontologies, such as autonomous vehicle benchmarks that refine a previous ${\tt vehicle}$ class into ${\tt school-bus}$ as autonomous operations expand to new cities. This paper formalizes a protocol for studying the problem of ${\it Learning with Evolving Class Ontology}$ (LECO). LECO requires learning classifiers in distinct time periods (TPs); each TP introduces a new ontology of "fine" labels that refines old ontologies of "coarse" labels (e.g., dog breeds that refine the previous ${\tt dog}$). LECO explores such questions as whether to annotate new data or relabel the old, how to exploit coarse labels, and whether to finetune the previous TP's model or train from scratch. To answer these questions, we leverage insights from related problems such as class-incremental learning. We validate them under the LECO protocol through the lens of image classification (on CIFAR and iNaturalist) and semantic segmentation (on Mapillary). Extensive experiments lead to some surprising conclusions; while the current status quo in the field is to relabel existing datasets with new class ontologies (such as COCO-to-LVIS or Mapillary1.2-to-2.0), LECO demonstrates that a far better strategy is to annotate ${\it new}$ data with the new ontology. However, this produces an aggregate dataset with inconsistent old-vs-new labels, complicating learning. To address this challenge, we adopt methods from semi-supervised and partial-label learning. We demonstrate that such strategies can surprisingly be made near-optimal, in the sense of approaching an "oracle" that learns on the aggregate dataset exhaustively labeled with the newest ontology. Zhiqiu Lin, Deepak Pathak, Yu-Xiong Wang, Deva Ramanan, Shu Kong	NeurIPS	2
2021	Worldsheet: Wrapping the World in a 3D Sheet for View Synthesis from a Single Image abstract We present Worldsheet, a method for novel view synthesis using just a single RGB image as input. The main insight is that simply shrink-wrapping a planar mesh sheet onto the input image, consistent with the learned intermediate depth, captures underlying geometry sufficient to generate photorealistic unseen views with large viewpoint changes. To operationalize this, we propose a novel differentiable texture sampler that allows our wrapped mesh sheet to be textured and rendered differentiably into an image from a target viewpoint. Our approach is category-agnostic, end-to-end trainable without using any 3D supervision, and requires a single image at test time. We also explore a simple extension by stacking multiple layers of Worldsheets to better handle occlusions. Worldsheet consistently outperforms prior state-of-the-art methods on single-image view synthesis across several datasets. Furthermore, this simple idea captures novel views surprisingly well on a wide range of high-resolution in-the-wild images, converting them into navigable 3D pop-ups. Video results and code are available at https://worldsheet.github.io. Ronghang Hu, Nikhila Ravi, Alexander C. Berg, Deepak Pathak	ICCV	4
2021	Learning Long-term Visual Dynamics with Region Proposal Interaction Networks Haozhi Qi, Xiaolong Wang 0004, Deepak Pathak, Yi Ma 0001, Jitendra Malik	ICLR	3
2021	Differentiable Spatial Planning using Transformers abstract We consider the problem of spatial path planning. In contrast to the classical solutions which optimize a new plan from scratch and assume access to the full map with ground truth obstacle locations, we learn a planner from the data in a differentiable manner that allows us to leverage statistical regularities from past data. We propose Spatial Planning Transformers (SPT), which given an obstacle map learns to generate actions by planning over long-range spatial dependencies, unlike prior data-driven planners that propagate information locally via convolutional structure in an iterative manner. In the setting where the ground truth map is not known to the agent, we leverage pre-trained SPTs in an end-to-end framework that has the structure of mapper and planner built into it which allows seamless generalization to out-of-distribution maps and goals. SPTs outperform prior state-of-the-art differentiable planners across all the setups for both manipulation and navigation tasks, leading to an absolute improvement of 7-19%. Devendra Singh Chaplot, Deepak Pathak, Jitendra Malik	ICML	2
2021	Unsupervised Learning of Visual 3D Keypoints for Control abstract Learning sensorimotor control policies from high-dimensional images crucially relies on the quality of the underlying visual representations. Prior works show that structured latent space such as visual keypoints often outperforms unstructured representations for robotic control. However, most of these representations, whether structured or unstructured are learned in a 2D space even though the control tasks are usually performed in a 3D environment. In this work, we propose a framework to learn such a 3D geometric structure directly from images in an end-to-end unsupervised manner. The input images are embedded into latent 3D keypoints via a differentiable encoder which is trained to optimize both a multi-view consistency loss and downstream task objective. These discovered 3D keypoints tend to meaningfully capture robot joints as well as object movements in a consistent manner across both time and 3D space. The proposed approach outperforms prior state-of-art methods across a variety of reinforcement learning benchmarks. Code and videos at https://buoyancy99.github.io/unsup-3d-keypoints/. Boyuan Chen 0003, Pieter Abbeel, Deepak Pathak	ICML	3
2021	Auto-Tuned Sim-to-Real Transfer abstract Policies trained in simulation often fail when transferred to the real world due to the ‘reality gap’ where the simulator is unable to accurately capture the dynamics and visual properties of the real world. Current approaches to tackle this problem, such as domain randomization, require prior knowledge and engineering to determine how much to randomize system parameters in order to learn a policy that is robust to sim-to-real transfer while also not being too conservative. We propose a method for automatically tuning simulator system parameters to match the real world using only raw RGB images of the real world without the need to define rewards or estimate state. Our key insight is to reframe the auto-tuning of parameters as a search problem where we iteratively shift the simulation system parameters to approach the real world system parameters. We propose a Search Param Model (SPM) that, given a sequence of observations and actions and a set of system parameters, predicts whether the given parameters are higher or lower than the true parameters used to generate the observations. We evaluate our method on multiple robotic control tasks in both sim-to-sim and sim-to-real transfer, demonstrating significant improvement over naive domain randomization. Project videos at https://yuqingd.github.io/autotuned-sim2real/. Olivia Watkins, Trevor Darrell, Pieter Abbeel, Deepak Pathak	ICRA	5
2021	Accelerating Robotic Reinforcement Learning via Parameterized Action Primitives abstract Despite the potential of reinforcement learning (RL) for building general-purpose robotic systems, training RL agents to solve robotics tasks still remains challenging due to the difficulty of exploration in purely continuous action spaces. Addressing this problem is an active area of research with the majority of focus on improving RL methods via better optimization or more efficient exploration. An alternate but important component to consider improving is the interface of the RL algorithm with the robot. In this work, we manually specify a library of robot action primitives (RAPS), parameterized with arguments that are learned by an RL policy. These parameterized primitives are expressive, simple to implement, enable efficient exploration and can be transferred across robots, tasks and environments. We perform a thorough empirical study across challenging tasks in three distinct domains with image input and a sparse terminal reward. We find that our simple change to the action interface substantially improves both the learning efficiency and task performance irrespective of the underlying RL algorithm, significantly outperforming prior methods which learn skills from offline expert data. Murtaza Dalal, Deepak Pathak, Ruslan Salakhutdinov	NeurIPS	2
2021	Functional Regularization for Reinforcement Learning via Learned Fourier Features abstract We propose a simple architecture for deep reinforcement learning by embedding inputs into a learned Fourier basis and show that it improves the sample efficiency of both state-based and image-based RL. We perform infinite-width analysis of our architecture using the Neural Tangent Kernel and theoretically show that tuning the initial variance of the Fourier basis is equivalent to functional regularization of the learned deep network. That is, these learned Fourier features allow for adjusting the degree to which networks underfit or overfit different frequencies in the training data, and hence provide a controlled mechanism to improve the stability and performance of RL optimization. Empirically, this allows us to prioritize learning low-frequency functions and speed up learning by reducing networks' susceptibility to noise in the optimization process, such as during Bellman updates. Experiments on standard state-based and image-based RL benchmarks show clear benefits of our architecture over the baselines. Alexander C. Li, Deepak Pathak	NeurIPS	2
2021	Discovering and Achieving Goals via World Models abstract How can artificial agents learn to solve many diverse tasks in complex visual environments without any supervision? We decompose this question into two challenges: discovering new goals and learning to reliably achieve them. Our proposed agent, Latent Explorer Achiever (LEXA), addresses both challenges by learning a world model from image inputs and using it to train an explorer and an achiever policy via imagined rollouts. Unlike prior methods that explore by reaching previously visited states, the explorer plans to discover unseen surprising states through foresight, which are then used as diverse targets for the achiever to practice. After the unsupervised phase, LEXA solves tasks specified as goal images zero-shot without any additional learning. LEXA substantially outperforms previous approaches to unsupervised goal reaching, both on prior benchmarks and on a new challenging benchmark with 40 test tasks spanning across four robotic manipulation and locomotion domains. LEXA further achieves goals that require interacting with multiple objects in sequence. Project page: https://orybkin.github.io/lexa/ Russell Mendonca, Oleh Rybkin, Kostas Daniilidis, Danijar Hafner, Deepak Pathak	NeurIPS	5
2021	Interesting Object, Curious Agent: Learning Task-Agnostic Exploration abstract Common approaches for task-agnostic exploration learn tabula-rasa --the agent assumes isolated environments and no prior knowledge or experience. However, in the real world, agents learn in many environments and always come with prior experiences as they explore new ones. Exploration is a lifelong process. In this paper, we propose a paradigm change in the formulation and evaluation of task-agnostic exploration. In this setup, the agent first learns to explore across many environments without any extrinsic goal in a task-agnostic manner.Later on, the agent effectively transfers the learned exploration policy to better explore new environments when solving tasks. In this context, we evaluate several baseline exploration strategies and present a simple yet effective approach to learning task-agnostic exploration policies. Our key idea is that there are two components of exploration: (1) an agent-centric component encouraging exploration of unseen parts of the environment based on an agent’s belief; (2) an environment-centric component encouraging exploration of inherently interesting objects. We show that our formulation is effective and provides the most consistent exploration across several training-testing environment pairs. We also introduce benchmarks and metrics for evaluating task-agnostic exploration strategies. The source code is available at https://github.com/sparisi/cbet/. Simone Parisi, Victoria Dean, Deepak Pathak, Abhinav Gupta 0001	NeurIPS	3
2020	The Origins of Common Sense in Humans and Machines Kevin A. Smith 0001, Eliza Kosoy, Alison Gopnik, Deepak Pathak, Alan Fern, Josh Tenenbaum, Tomer D. Ullman	CogSci	4
2020	Exploring Exploration: Comparing Children with Agents in Unified Exploration Environments Eliza Kosoy, Jasmine Collins, David M. Chan, Deepak Pathak, Pulkit Agrawal 0001, Alison Gopnik	CogSci	4
2020	Tangible Flowchart Blocks for Fostering Logical Thinking in Visually Impaired Learners abstract Logical thinking is an essential skill needed to gain expertise in programming. Typically logical thinking is learned using a flowchart, which is a step by step process to reach a particular solution. Because of the visual constraint, teaching logical thinking to visually impaired (VI) students becomes challenging. For VI students` concrete experience with real objects is necessary for creating memory. For VI students haptic learning can be facilitated by designing flowchart blocks with appropriate feedback. All the design considerations are made considering the interaction style or challenges for VI students. We focus on developing tangible flowchart blocks to foster logical thinking in VI students. Learning activities were designed using contextual problems where VI students interact with the blocks to form a step-by-step solution. In this paper, we present findings from a pilot study and interviews conducted with three participants for early validation and feedback for the scope of improvement of our product. Priya C. Lotlikar, Deepak Pathak, Herold P. C., Chandan Dasgupta	ICALT	2
2020	Impact of Gender on Motivation, Engagement and Interaction Behavior in Mobile assisted learning of English abstract In developing countries like India, proficiency in English is a desired skill since it boosts job prospects and improves economic condition. However, students' proficiency level remains inadequate, even though English is taught in schools. Mobile language learning apps like Hello English addresses this gap by providing learners with interactive activities, collaborative learning opportunities and game-based challenges to develop their English proficiency. In this context, the current paper explores three research questions that examines gender-based variation in motivation, engagement and interaction behavior of school students who use Hello English. Students' motivation level and their subsequent engagement with the app are measured through two separate questionnaires. Gender difference in students' interaction behavior is analyzed using their interaction data with the app. We found that female students are highly motivated, prefers to interact with the app more often and spend more time on app compared to male students. Ramkumar Rajendran, Gargi Banerjee, Deepak Pathak, Sivaranjani Sivamohan	ICALT	3
2020	One Policy to Control Them All: Shared Modular Policies for Agent-Agnostic Control abstract Reinforcement learning is typically concerned with learning control policies tailored to a particular agent. We investigate whether there exists a single global policy that can generalize to control a wide variety of agent morphologies – ones in which even dimensionality of state and action spaces changes. We propose to express this global policy as a collection of identical modular neural networks, dubbed as Shared Modular Policies (SMP), that correspond to each of the agent’s actuators. Every module is only responsible for controlling its corresponding actuator and receives information from only its local sensors. In addition, messages are passed between modules, propagating information between distant modules. We show that a single modular policy can successfully generate locomotion behaviors for several planar agents with different skeletal structures such as monopod hoppers, quadrupeds, bipeds, and generalize to variants not seen during training – a process that would normally require training and manual hyperparameter tuning for each morphology. We observe that a wide variety of drastically diverse locomotion styles across morphologies as well as centralized coordination emerges via message passing between decentralized modules purely from the reinforcement learning objective. Videos and code at https://huangwl18.github.io/modular-rl/ Wenlong Huang, Igor Mordatch, Deepak Pathak	ICML	3
2020	Planning to Explore via Self-Supervised World Models abstract Reinforcement learning allows solving complex tasks, however, the learning tends to be task-specific and the sample efficiency remains a challenge. We present Plan2Explore, a self-supervised reinforcement learning agent that tackles both these challenges through a new approach to self-supervised exploration and fast adaptation to new tasks, which need not be known during exploration. During exploration, unlike prior methods which retrospectively compute the novelty of observations after the agent has already reached them, our agent acts efficiently by leveraging planning to seek out expected future novelty. After exploration, the agent quickly adapts to multiple downstream tasks in a zero or a few-shot manner. We evaluate on challenging control tasks from high-dimensional image inputs. Without any training supervision or task-specific interaction, Plan2Explore outperforms prior self-supervised exploration methods, and in fact, almost matches the performances oracle which has access to rewards. Videos and code: https://ramanans1.github.io/plan2explore/ Ramanan Sekar, Oleh Rybkin, Kostas Daniilidis, Pieter Abbeel, Danijar Hafner, Deepak Pathak	ICML	6
2020	Neural Dynamic Policies for End-to-End Sensorimotor Learning abstract The current dominant paradigm in sensorimotor control, whether imitation or reinforcement learning, is to train policies directly in raw action spaces such as torque, joint angle, or end-effector position. This forces the agent to make decision at each point in training, and hence, limits the scalability to continuous, high-dimensional, and long-horizon tasks. In contrast, research in classical robotics has, for a long time, exploited dynamical systems as a policy representation to learn robot behaviors via demonstrations. These techniques, however, lack the flexibility and generalizability provided by deep learning or deep reinforcement learning and have remained under-explored in such settings. In this work, we begin to close this gap and embed dynamics structure into deep neural network-based policies by reparameterizing action spaces with differential equations. We propose Neural Dynamic Policies (NPDs) that make predictions in trajectory distribution space as opposed to prior policy learning methods where action represents the raw control space. The embedded structure allows us to perform end-to-end policy learning under both reinforcement and imitation learning setups. We show that NDPs achieve better or comparable performance to state-of-the-art approaches on many robotic control tasks using both reward-based training and demonstrations. Project video and code are available at: https://shikharbahl.github.io/neural-dynamic-policies/. Shikhar Bahl, Mustafa Mukadam, Abhinav Gupta 0001, Deepak Pathak	NeurIPS	4
2020	Sparse Graphical Memory for Robust Planning abstract To operate effectively in the real world, agents should be able to act from high-dimensional raw sensory input such as images and achieve diverse goals across long time-horizons. Current deep reinforcement and imitation learning methods can learn directly from high-dimensional inputs but do not scale well to long-horizon tasks. In contrast, classical graphical methods like A* search are able to solve long-horizon tasks, but assume that the state space is abstracted away from raw sensory input. Recent works have attempted to combine the strengths of deep learning and classical planning; however, dominant methods in this domain are still quite brittle and scale poorly with the size of the environment. We introduce Sparse Graphical Memory (SGM), a new data structure that stores states and feasible transitions in a sparse memory. SGM aggregates states according to a novel two-way consistency objective, adapting classic state aggregation criteria to goal-conditioned RL: two states are redundant when they are interchangeable both as goals and as starting states. Theoretically, we prove that merging nodes according to two-way consistency leads to an increase in shortest path lengths that scales only linearly with the merging threshold. Experimentally, we show that SGM significantly outperforms current state of the art methods on long horizon, sparse-reward visual navigation tasks. Project video and code are available at https://sites.google.com/view/sparse-graphical-memory. Scott Emmons, Ajay Jain, Michael Laskin, Thanard Kurutach, Pieter Abbeel, Deepak Pathak	NeurIPS	6
2020	Locally Masked Convolution for Autoregressive Models abstract High-dimensional generative models have many applications including image compression, multimedia generation, anomaly detection and data completion. State-of-the-art estimators for natural images are autoregressive, decomposing the joint distribution over pixels into a product of conditionals parameterized by a deep neural network, e.g. a convolutional neural network such as the PixelCNN. However, PixelCNNs only model a single decomposition of the joint, and only a single generation order is efficient. For tasks such as image completion, these models are unable to use much of the observed context. To generate data in arbitrary orders, we introduce LMConv: a simple modification to the standard 2D convolution that allows arbitrary masks to be applied to the weights at each location in the image. Using LMConv, we learn an ensemble of distribution estimators that share parameters but differ in generation order, achieving improved performance on whole-image density estimation (2.89 bpd on unconditional CIFAR10), as well as globally coherent image completions. Code is available at https://ajayjain.github.io/lmconv. Ajay Jain, Pieter Abbeel, Deepak Pathak	UAI	3
2020	Compositional GAN: Learning Image-Conditional Binary Composition Samaneh Azadi, Deepak Pathak, Sayna Ebrahimi, Trevor Darrell	Int. J. Comput. Vis.	2
2019	Human-level but not human-like: Deep Reinforcement Learning in the dark Rachit Dubey, Pulkit Agrawal 0001, Deepak Pathak, Alexei A. Efros, Thomas L. Griffiths 0001	CogSci	3
2019	Curiouser and Curiouser: Children's intrinsic exploration of mazes and its effects on reaching a goal Eliza Kosoy, Deepak Pathak, Pulkit Agrawal 0001, Alison Gopnik	CogSci	2
2019	Large-Scale Study of Curiosity-Driven Learning Yuri Burda, Deepak Pathak, Amos J. Storkey, Trevor Darrell, Alexei A. Efros	ICLR (Poster)	3
2019	Self-Supervised Exploration via Disagreement abstract Efficient exploration is a long-standing problem in sensorimotor learning. Major advances have been demonstrated in noise-free, non-stochastic domains such as video games and simulation. However, most of these formulations either get stuck in environments with stochastic dynamics or are too inefficient to be scalable to real robotics setups. In this paper, we propose a formulation for exploration inspired by the work in active learning literature. Specifically, we train an ensemble of dynamics models and incentivize the agent to explore such that the disagreement of those ensembles is maximized. This allows the agent to learn skills by exploring in a self-supervised manner without any external reward. Notably, we further leverage the disagreement objective to optimize the agent’s policy in a differentiable manner, without using reinforcement learning, which results in a sample-efficient exploration. We demonstrate the efficacy of this formulation across a variety of benchmark environments including stochastic-Atari, Mujoco and Unity. Finally, we implement our differentiable exploration on a real robot which learns to interact with objects completely from scratch. Project videos and code are at https://pathak22.github.io/exploration-by-disagreement/ Deepak Pathak, Dhiraj Gandhi, Abhinav Gupta 0001	ICML	1
2019	Learning to Control Self-Assembling Morphologies: A Study of Generalization via Modularity abstract Contemporary sensorimotor learning approaches typically start with an existing complex agent (e.g., a robotic arm), which they learn to control. In contrast, this paper investigates a modular co-evolution strategy: a collection of primitive agents learns to dynamically self-assemble into composite bodies while also learning to coordinate their behavior to control these bodies. Each primitive agent consists of a limb with a motor attached at one end. Limbs may choose to link up to form collectives. When a limb initiates a link-up action and there is another limb nearby, the latter is magnetically connected to the 'parent' limb's motor. This forms a new single agent, which may further link with other agents. In this way, complex morphologies can emerge, controlled by a policy whose architecture is in explicit correspondence with the morphology. We evaluate the performance of these dynamic and modular agents in simulated environments. We demonstrate better generalization to test-time changes both in the environment, as well as in the structure of the agent, compared to static and monolithic baselines. Project videos and source code are provided in the supplementary material. Deepak Pathak, Chris Lu 0001, Trevor Darrell, Phillip Isola, Alexei A. Efros	NeurIPS	1
2019	Third-Person Visual Imitation Learning via Decoupled Hierarchical Controller abstract We study a generalized setup for learning from demonstration to build an agent that can manipulate novel objects in unseen scenarios by looking at only a single video of human demonstration from a third-person perspective. To accomplish this goal, our agent should not only learn to understand the intent of the demonstrated third-person video in its context but also perform the intended task in its environment configuration. Our central insight is to enforce this structure explicitly during learning by decoupling what to achieve (intended task) from how to perform it (controller). We propose a hierarchical setup where a high-level module learns to generate a series of first-person sub-goals conditioned on the third-person video demonstration, and a low-level controller predicts the actions to achieve those sub-goals. Our agent acts from raw image observations without any access to the full state information. We show results on a real robotic platform using Baxter for the manipulation tasks of pouring and placing objects in a box. Project video is available at https://pathak22.github.io/hierarchical-imitation/ Pratyusha Sharma, Deepak Pathak, Abhinav Gupta 0001	NeurIPS	2
2018	Zero-Shot Visual Imitation Deepak Pathak, Parsa Mahmoudieh, Guanghao Luo, Pulkit Agrawal 0001, Dian Chen 0001, Yide Shentu, Evan Shelhamer, Jitendra Malik, Alexei A. Efros, Trevor Darrell	ICLR	1
2018	Investigating Human Priors for Playing Video Games abstract What makes humans so good at solving seemingly complex video games? Unlike computers, humans bring in a great deal of prior knowledge about the world, enabling efficient decision making. This paper investigates the role of human priors for solving video games. Given a sample game, we conduct a series of ablation studies to quantify the importance of various priors on human performance. We do this by modifying the video game environment to systematically mask different types of visual information that could be used by humans as priors. We find that removal of some prior knowledge causes a drastic degradation in the speed with which human players solve the game, e.g. from 2 minutes to over 20 minutes. Furthermore, our results indicate that general priors, such as the importance of objects and visual consistency, are critical for efficient game-play. Videos and the game manipulations are available at https://rach0012.github.io/humanRL_website/ Rachit Dubey, Pulkit Agrawal 0001, Deepak Pathak, Thomas L. Griffiths 0001, Alexei A. Efros	ICML	3
2017	Learning Features by Watching Objects Move abstract This paper presents a novel yet intuitive approach to unsupervised feature learning. Inspired by the human visual system, we explore whether low-level motion-based grouping cues can be used to learn an effective visual representation. Specifically, we use unsupervised motion-based segmentation on videos to obtain segments, which we use as pseudo ground truth to train a convolutional network to segment objects from a single frame. Given the extensive evidence that motion plays a key role in the development of the human visual system, we hope that this straightforward approach to unsupervised learning will be more effective than cleverly designed pretext tasks studied in the literature. Indeed, our extensive experiments show that this is the case. When used for transfer learning on object detection, our representation significantly outperforms previous unsupervised approaches across multiple settings, especially when training data for the target task is scarce. Deepak Pathak, Ross B. Girshick, Piotr Dollár, Trevor Darrell, Bharath Hariharan	CVPR	1
2017	Curiosity-driven Exploration by Self-supervised Prediction abstract In many real-world scenarios, rewards extrinsic to the agent are extremely sparse, or absent altogether. In such cases, curiosity can serve as an intrinsic reward signal to enable the agent to explore its environment and learn skills that might be useful later in its life. We formulate curiosity as the error in an agent’s ability to predict the consequence of its own actions in a visual feature space learned by a self-supervised inverse dynamics model. Our formulation scales to high-dimensional continuous state spaces like images, bypasses the difficulties of directly predicting pixels, and, critically, ignores the aspects of the environment that cannot affect the agent. The proposed approach is evaluated in two environments: VizDoom and Super Mario Bros. Three broad settings are investigated: 1) sparse extrinsic reward, where curiosity allows for far fewer interactions with the environment to reach the goal; 2) exploration with no extrinsic reward, where curiosity pushes the agent to explore more efficiently; and 3) generalization to unseen scenarios (e.g. new levels of the same game) where the knowledge gained from earlier experience helps the agent explore new places much faster than starting from scratch. Deepak Pathak, Pulkit Agrawal 0001, Alexei A. Efros, Trevor Darrell	ICML	1
2017	Toward Multimodal Image-to-Image Translation abstract Many image-to-image translation problems are ambiguous, as a single input image may correspond to multiple possible outputs. In this work, we aim to model a distribution of possible outputs in a conditional generative modeling setting. The ambiguity of the mapping is distilled in a low-dimensional latent vector, which can be randomly sampled at test time. A generator learns to map the given input, combined with this latent code, to the output. We explicitly encourage the connection between output and the latent code to be invertible. This helps prevent a many-to-one mapping from the latent code to the output during training, also known as the problem of mode collapse, and produces more diverse results. We explore several variants of this approach by employing different training objectives, network architectures, and methods of injecting the latent code. Our proposed method encourages bijective consistency between the latent encoding and output modes. We present a systematic comparison of our method and other variants on both perceptual realism and diversity. Jun-Yan Zhu, Richard Zhang 0001, Deepak Pathak, Trevor Darrell, Alexei A. Efros, Oliver Wang, Eli Shechtman	NIPS	3
2016	Context Encoders: Feature Learning by Inpainting abstract We present an unsupervised visual feature learning algorithm driven by context-based pixel prediction. By analogy with auto-encoders, we propose Context Encoders - a convolutional neural network trained to generate the contents of an arbitrary image region conditioned on its surroundings. In order to succeed at this task, context encoders need to both understand the content of the entire image, as well as produce a plausible hypothesis for the missing part(s). When training context encoders, we have experimented with both a standard pixel-wise reconstruction loss, as well as a reconstruction plus an adversarial loss. The latter produces much sharper results because it can better handle multiple modes in the output. We found that a context encoder learns a representation that captures not just appearance but also the semantics of visual structures. We quantitatively demonstrate the effectiveness of our learned features for CNN pre-training on classification, detection, and segmentation tasks. Furthermore, context encoders can be used for semantic inpainting tasks, either stand-alone or as initialization for non-parametric methods. Deepak Pathak, Philipp Krähenbühl, Jeff Donahue, Trevor Darrell, Alexei A. Efros	CVPR	1
2016	Large Scale Visual Recognition through Adaptation using Joint Representation and Multiple Instance Learning abstract A major barrier towards scaling visual recognition systems is the difficulty of obtaining labeled images for large numbers of categories. Recently, deep convolutional neural networks (CNNs) trained used 1.2M+ labeled images have emerged as clear winners on object classification benchmarks. Unfortunately, only a small fraction of those labels are available with bounding box localization for training the detection task and even fewer pixel level annotations are available for semantic segmentation. It is much cheaper and easier to collect large quantities of image-level labels from search engines than it is to collect scene-centric images with precisely localized labels. We develop methods for learning large scale recognition models which exploit joint training over both weak (image-level) and strong (bounding box) labels and which transfer learned perceptual representations from strongly-labeled auxiliary tasks. We provide a novel formulation of a joint multiple instance learning method that includes examples from object-centric data with image-level labels when available, and also performs domain transfer learning to improve the underlying detector representation. We then show how to use our large scale detectors to produce pixel level annotations. Using our method, we produce a $>$7.6K category detector and release code and models at lsda.berkeley vision.org. Judy Hoffman, Deepak Pathak, Eric Tzeng, Jonathan Long, Sergio Guadarrama, Trevor Darrell, Kate Saenko	J. Mach. Learn. Res.	2
2015	Detector discovery in the wild: Joint multiple instance and representation learning abstract We develop methods for detector learning which exploit joint training over both weak (image-level) and strong (bounding box) labels and which transfer learned perceptual representations from strongly-labeled auxiliary tasks. Previous methods for weak-label learning often learn detector models independently using latent variable optimization, but fail to share deep representation knowledge across classes and usually require strong initialization. Other previous methods transfer deep representations from domains with strong labels to those with only weak labels, but do not optimize over individual latent boxes, and thus may miss specific salient structures for a particular category. We propose a model that subsumes these previous approaches, and simultaneously trains a representation and detectors for categories with either weak or strong labels present. We provide a novel formulation of a joint multiple instance learning method that includes examples from classification-style data when available, and also performs domain transfer learning to improve the underlying detector representation. Our model outperforms known methods on ImageNet-200 detection with weak labels. Judy Hoffman, Deepak Pathak, Trevor Darrell, Kate Saenko	CVPR	2
2015	Constrained Convolutional Neural Networks for Weakly Supervised Segmentation abstract We present an approach to learn a dense pixel-wise labeling from image-level tags. Each image-level tag imposes constraints on the output labeling of a Convolutional Neural Network (CNN) classifier. We propose Constrained CNN (CCNN), a method which uses a novel loss function to optimize for any set of linear constraints on the output space (i.e. predicted label distribution) of a CNN. Our loss formulation is easy to optimize and can be incorporated directly into standard stochastic gradient descent optimization. The key idea is to phrase the training objective as a biconvex optimization for linear models, which we then relax to nonlinear deep networks. Extensive experiments demonstrate the generality of our new learning framework. The constrained loss yields state-of-the-art results on weakly supervised semantic image segmentation. We further demonstrate that adding slightly more supervision can greatly improve the performance of the learning algorithm. Deepak Pathak, Philipp Krähenbühl, Trevor Darrell	ICCV	1
2015	Anomaly Localization in Topic-Based Analysis of Surveillance Videos abstract Topic-models for video analysis have been used for unsupervised identification of normal activity in videos, thereby enabling the detection of anomalous actions. However, while intervals containing anomalies are detected, it has not been possible to localize the anomalous activities in such models. This is a challenging problem as the abnormal content is usually a small fraction of the entire video data and hence distinctions in terms of likelihood are unlikely. Here we propose a methodology to extend the topic based analysis with rich local descriptors incorporating quantized spatio-temporal gradient descriptors with image location and size information. The visual clips over this vocabulary are then represented in latent topic space using models like pLSA. Further, we introduce an algorithm to quantify the anomalous content in a video clip by projecting the learned topic space information. Using the algorithm, we detect whether the video clip is abnormal and if positive, localize the anomaly in spatio-temporal domain. We also contribute one real world surveillance video dataset for comprehensive evaluation of the proposed algorithm. Experiments are presented on the proposed and two other standard surveillance datasets. Deepak Pathak, Abhijit Sharang, Amitabha Mukerjee	WACV	1