Imitation learning (IL) is a well-known problem in the field of Markov decision process (MDP), where one is given multiple demonstration trajectories generated by expert(s), and the goal is to replicate the hidden expert-policies so that when the MDP is run independently, it generates trajectories close to the demonstrated ones. IL is one of the most useful tools used in building versatile robots that can learn from examples. This task becomes particularly challenging when the expert exhibits a mixture of behavior modes. Prior work has introduced latent variables to model variations of the expert policy. However, our experiments show that the existing works do not exhibit appropriate imitation of individual modes. To tackle this problem, we first draw inspiration from the well-known classical technique of self-organizing maps (SOMs) and introduce an encoder-free generative model-referred to as the self-organizing generative (SOG) model-for learning multimodal data distributions from samples. We then apply SOG for behavior cloning (BC)-a framework that learns deterministic policies without considering the environment-to accurately distinguish and imitate different modes. Then, we integrate it with generative adversarial IL (GAIL)-a framework that learns policies while considering the environment-to make the learning robust toward compounding errors at unseen states. We show that our method significantly outperforms the state of the art across multiple experiments within the MuJoCo simulator, including locomotion and robotic manipulation tasks.

Predicting future trajectories of road agents is a critical task for autonomous driving. Recent goal-based trajectory prediction methods, such as DenseTNT and PECNet [1], [2], have shown good performance on prediction tasks on public datasets. However, they usually require complicated goal-selection algorithms and optimization. In this work, we propose KEMP, a hierarchical end-to-end deep learning framework for trajectory prediction. At the core of our framework is keyframe-based trajectory prediction, where keyframes are representative states that trace out the general direction of the trajectory. KEMP first predicts keyframes conditioned on the road con-text, and then fills in intermediate states conditioned on the keyframes and the road context. Under our general framework, goal-conditioned methods are special cases in which the number of keyframes equal to one. Unlike goal-conditioned methods, our keyframe predictor is learned automatically and does not require hand-crafted goal-selection algorithms. We evaluate our model on public benchmarks and our model ranked 1st on Waymo Open Motion Dataset Leaderboard (as of September 1, 2021).

We propose a novel framework, On-Demand MOtion Generation (ODMO), for generating realistic and diverse long-term 3D human motion sequences conditioned only on action types with an additional capability of customization. ODMO shows improvements over SOTA approaches on all traditional motion evaluation metrics when evaluated on three public datasets (HumanAct12, UESTC, and MoCap). Furthermore, we provide both qualitative evaluations and quantitative metrics demonstrating several first-known customization capabilities afforded by our framework, including mode discovery, interpolation, and trajectory customization. These capabilities significantly widen the spectrum of potential applications of such motion generation models. The novel on-demand generative capabilities are enabled by innovations in both the encoder and decoder architectures: (i) Encoder: Utilizing contrastive learning in low-dimensional latent space to create a hierarchical embedding of motion sequences, where not only the codes of different action types form different groups, but within an action type, codes of similar inherent patterns (motion styles) cluster together, making them readily discoverable; (ii) Decoder: Using a hierarchical decoding strategy where the motion trajectory is reconstructed first and then used to reconstruct the whole motion sequence. Such an architecture enables effective trajectory control. Our code is released on the Github page: https://github.com/roychowdhuryresearch/ODMO https://github.com/roychowdhuryresearch/ODMO

In advertising, identifying the content safety of web pages is a significant concern since advertisers do not want brands to be associated with threatening content. At the same time, publishers would like to maximize the number of web pages on which they can place ads. Thus, a fine balance must be achieved while classifying content safety in order to satisfy both advertisers and publishers. In this paper, we propose a multimodal machine learning framework that fuses visual and textual information from web pages to improve current predictions of content safety. The primary focus is on late fusion, which involves combining final model outputs of separate modalities, such as images and text, to arrive at a single decision. This paper presents a fully automated machine learning framework that performs binary and multilabel classification using late fusion techniques. We also introduce additional work in early fusion, which involves extracting and fusing intermediate features from the two separate models. Our algorithms are applied to data extracted from relevant web pages in the advertising industry. Both of our late and early fusion methods obtain significant improvements over algorithms currently in use.