Computational complexity of Bayesian learning is impeding its adoption in practical, large-scale tasks, despite demonstrations of significant merits such as improved robustness and resilience to unseen or out-of-distribution inputs over their non-Bayesian counterparts. Although, Deep ensemble methods (Seligmann et al. 2024; Lakshminarayanan, Pritzel, and Blundell 2017) have proven to be highly effective for Bayesian deep learning, their practical application is hindered by substantial computational cost. In this study, we introduce an innovative framework to mitigate the computational burden of ensemble Bayesian deep learning. We explore a more feasible alternative, inspired by the recent success of low-rank adapters, we introduce Bayesian Low-Rank LeArning (Bella). We show, i) Bella achieves a dramatic reduction in the number of trainable parameters required to approximate a Bayesian posterior; and ii) it not only maintains, but in some instances, surpasses the performance–in accuracy and out-of-distribution generalisation–of conventional Bayesian learning methods and non-Bayesian baselines. Our extensive empirical evaluation in large-scale tasks such as ImageNet, CAMELYON17, DomainNet, VQA with CLIP, LLaVA demonstrate the effectiveness and versatility of Bella in building highly scalable and practical Bayesian deep models for real-world applications.

Neural architectures tend to fit their data with relatively simple functions. This "simplicity bias" is widely regarded as key to their success. This paper explores the limits of this principle. Building on recent findings that the simplicity bias stems from ReLU activations [96], we introduce a method to meta-learn new activation functions and inductive biases better suited to specific tasks.Findings. We identify multiple tasks where the simplicity bias is inadequate and ReLUs suboptimal. In these cases, we learn new activation functions that perform better by inducing a prior of higher complexity. Interestingly, these cases correspond to domains where neural networks have historically struggled: tabular data, regression tasks, cases of shortcut learning, and algorithmic grokking tasks. In comparison, the simplicity bias induced by ReLUs proves adequate on image tasks where the best learned activations are nearly identical to ReLUs and GeLUs.Implications. Contrary to popular belief, the simplicity bias of ReLU networks is not universally useful. It is near-optimal for image classification, but other inductive biases are sometimes preferable. We showed that activation functions can control these inductive biases, but future tailored architectures might provide further benefits. Advances are still needed to characterize a model’s inductive biases beyond "complexity", and their adequacy with the data.

Contrastive Language-Image Pretraining (CLIP) stands out as a prominent method for image representation learning. Various architectures, from vision transformers~(ViTs) to convolutional networks (ResNets) have been trained with CLIP to serve as general solutions to diverse vision tasks. This paper explores the differences across various CLIP-trained vision backbones. Despite using the same data and training objective, we find that these architectures have notably different representations, different classification performance across datasets, and different robustness properties to certain types of image perturbations. Our findings indicate a remarkable possible synergy across backbones by leveraging their respective strengths. In principle, classification accuracy could be improved by over 40 percentage with an informed selection of the optimal backbone per test example. Using this insight, we develop a straightforward yet powerful approach to adaptively ensemble multiple backbones. The approach uses as few as one labeled example per class to tune the adaptive combination of backbones. On a large collection of datasets, the method achieves a remarkable increase in accuracy of up to 39.1\% over the best single backbone, well beyond traditional ensembles.

Out-of-distribution (OOD) generalization is challenging because distribution shifts come in many forms. Numerous algorithms exist to address specific settings, but *choosing the right training algorithm for the right dataset* without trial and error is difficult. Indeed, real-world applications often involve multiple types and combinations of shifts that are hard to analyze theoretically. **Method.** This work explores the possibility of *learning* the selection of a training algorithm for OOD generalization. We propose a proof of concept (OOD-Chameleon) that formulates the selection as a multi-label classification over candidate algorithms, trained on a *dataset of datasets* representing a variety of shifts. We evaluate the ability of OOD-Chameleon to rank algorithms on unseen shifts and datasets based only on dataset characteristics, i.e., without training models first, unlike traditional model selection. **Findings.** Extensive experiments show that the learned selector identifies high-performing algorithms across synthetic, vision, and language tasks. Further inspection shows that it learns non-trivial decision rules, which provide new insights into the applicability of existing algorithms. Overall, this new approach opens the possibility of better exploiting and understanding the plethora of existing algorithms for OOD generalization.

Out-of-distribution (OOD) detection is essential for reliably deploying machine learning models in the wild. Yet, most methods treat large pre-trained models as monolithic encoders and rely solely on their final-layer representations for detection. We challenge this wisdom. We reveal the intermediate layers of pre-trained models, shaped by residual connections that subtly transform input projections, can encode surprisingly rich and diverse signals for detecting distributional shifts. Importantly, to exploit latent representation diversity across layers, we introduce an entropy-based criterion to automatically identify layers offering the most complementary information in a training-free setting, without access to OOD data. We show that selectively incorporating these intermediate representations can increase the accuracy of OOD detection by up to $10\%$ in far-OOD and over $7\%$ in near-OOD benchmarks compared to state-of-the-art training-free methods across various model architectures and training objectives. Our findings reveal a new avenue for OOD detection research and uncover the impact of various training objectives and model architectures on confidence-based OOD detection methods.

The increasing reliance on large-scale datasets in machine learning poses significant privacy and ethical challenges, particularly in sensitive domains such as face recognition. Synthetic data generation offers a promising alternative; however, most existing methods depend heavily on external datasets or pre-trained models, increasing complexity and resource demands. In this paper, we introduce **AugGen**, a self-contained synthetic augmentation technique. AugGen strategically samples from a class-conditional generative model trained exclusively on the target FR dataset, eliminating the need for external resources. Evaluated across 8 FR benchmarks, including IJB-C and IJB-B, our method achieves **1–12% performance improvements**, outperforming models trained solely on real data and surpassing state-of-the-art synthetic data generation approaches, while using less real data. Notably, these gains often exceed those from architectural modifications, underscoring the value of synthetic augmentation in data-limited scenarios. Our findings demonstrate that carefully integrated synthetic data can both mitigate privacy constraints and substantially enhance discriminative performance in face recognition. Code and datasets will be made publicly available upon publication.

Our understanding of the generalization capabilities of neural networks (NNs) is still incomplete. Prevailing explanations are based on implicit biases of gradient de-scent (GD) but they cannot account for the capabilities of models from gradient-free methods [9] nor the simplicity bias recently observed in untrained networks [29]. This paper seeks other sources of generalization in NNs. Findings. To understand the inductive biases provided by architectures independently from GD, we examine un-trained, random-weight networks. Even simple MLPs show strong inductive biases: uniform sampling in weight space yields a very biased distribution of functions in terms of complexity. But unlike common wisdom, NNs do not have an inherent “simplicity bias”. This property depends on components such as ReLUs, residual connections, and layer normalizations. Alternative architectures can be built with a bias for any level of complexity. Transformers also inherit all these properties from their building blocks. Implications. We provide a fresh explanation for the success of deep learning independent from gradient-based training. It points at promising avenues for controlling the solutions implemented by trained models.

Mixup is a highly successful technique to improve generalization by augmenting training data with combinations of random pairs. Selective mixup is a family of methods that apply mixup to specific pairs e.g. combining examples across classes or domains. Despite remarkable performance on benchmarks with distribution shifts, these methods are still poorly understood. We find that an overlooked aspect of selective mixup explains some of its success in a completely new light. The non-random selection of pairs affects the training distribution and improves generalization by means completely unrelated to the mixing. For example in binary classification, mixup across classes implicitly resamples the data to uniform class distribution - a classical solution to label shift. We verify empirically that this resampling explains some of the improvements reported in prior work. Theoretically, the effect relies on a “regression toward the mean”, an accidental property we find in several datasets. Outcomes. We now better understand why selective mixup works. This lets us predict a yet-unknown failure mode and conditions where the method is detrimental. We also use the equivalence with resampling to design better variants that combine mixing and resampling effects.

Jordan Meadows, Marco Valentino, Damien Teney, Andre Freitas. Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2024.

Cultural bias is pervasive in many large language models (LLMs), largely due to the deficiency of data representative of different cultures. Typically, cultural datasets and benchmarks are constructed either by extracting subsets of existing datasets or by aggregating from platforms such as Wikipedia and social media. However, these approaches are highly dependent on real-world data and human annotations, making them costly and difficult to scale. Inspired by cognitive theories on social communication, this paper introduces CulturePark, an LLM-powered multi-agent communication framework for cultural data collection. CulturePark simulates cross-cultural human communication with LLM-based agents playing roles in different cultures. It generates high-quality cross-cultural dialogues encapsulating human beliefs, norms, and customs. Using CulturePark, we generated 41,000 cultural samples to fine-tune eight culture-specific LLMs. We evaluated these models across three downstream tasks: content moderation, cultural alignment, and cultural education. Results show that for content moderation, our GPT-3.5-based models either match or outperform GPT-4 on $41$ datasets. Regarding cultural alignment, our models surpass GPT-4 on Hofstede's VSM 13 framework. Furthermore, for cultural education of human participants, our models demonstrate superior outcomes in both learning efficacy and user experience compared to GPT-4. CulturePark proves an important step in addressing cultural bias and advancing the democratization of AI, highlighting the critical role of culturally inclusive data in model training. Code is released at https://github.com/Scarelette/CulturePark.

Composed image retrieval searches for a target image based on a multi-modal user query comprised of a reference image and modification text describing the desired changes. Existing approaches to solving this challenging task learn a mapping from the (reference image, modification text)-pair to an image embedding that is then matched against a large image corpus. One area that has not yet been explored is the reverse direction, which asks the question, what reference image when modified as described by the text would produce the given target image? In this work we propose a bi-directional training scheme that leverages such reversed queries and can be applied to existing composed image retrieval architectures with minimum changes, which improves the performance of the model. To encode the bi-directional query we prepend a learnable token to the modification text that designates the direction of the query and then finetune the parameters of the text embedding module. We make no other changes to the network architecture. Experiments on two standard datasets show that our novel approach achieves improved performance over a baseline BLIP-based model that itself already achieves competitive performance. Our code is released at https://github.com/Cuberick-Orion/Bi-Blip4CIR

Several studies have compared the in-distribution (ID) and out-of-distribution (OOD) performance of models in computer vision and NLP. They report a frequent positive correlation and some surprisingly never even observe an inverse correlation indicative of a necessary trade-off. The possibility of inverse patterns is important to determine whether ID performance can serve as a proxy for OOD generalization capabilities. This paper shows that inverse correlations between ID and OOD performance do happen with multiple real-world datasets, not only in artificial worst-case settings. We explain theoretically how these cases arise and how past studies missed them because of improper methodologies that examined a biased selection of models. Our observations lead to recommendations that contradict those found in much of the current literature. - High OOD performance sometimes requires trading off ID performance. - Focusing on ID performance alone may not lead to optimal OOD performance. It may produce diminishing (eventually negative) returns in OOD performance. - In these cases, studies on OOD generalization that use ID performance for model selection (a common recommended practice) will necessarily miss the best-performing models, making these studies blind to a whole range of phenomena.

The promise of active learning (AL) is to reduce labelling costs by selecting the most valuable examples to annotate from a pool of unlabelled data. Identifying these examples is especially challenging with high-dimensional data (e.g. images, videos) and in low-data regimes. In this paper, we propose a novel method for batch AL called ALFA-Mix. We identify unlabelled instances with sufficiently-distinct features by seeking inconsistencies in predictions resulting from interventions on their representations. We construct interpolations between representations of labelled and unlabelled instances then examine the predicted labels. We show that inconsistencies in these predictions help discovering features that the model is unable to recognise in the unlabelled instances. We derive an efficient implementation based on a closed-form solution to the optimal interpolation causing changes in predictions. Our method outperforms all recent AL approaches in 30 different settings on 12 benchmarks of images, videos, and non-visual data. The improvements are especially significant in low-data regimes and on self-trained vision transformers, where ALFA-Mix outperforms the state-of-the-art in 59% and 43% of the experiments respectively11The code is available at https://github.com/aminparvaneh/alpha_mix_active_learning.

Neural networks trained with SGD were recently shown to rely preferentially on linearly-predictive features and can ignore complex, equally-predictive ones. This simplicity bias can explain their lack of robustness out of distribution (OOD). The more complex the task to learn, the more likely it is that statistical artifacts (i.e. selection biases, spurious correlations) are simpler than the mechanisms to learn. We demonstrate that the simplicity bias can be mitigated and OOD generalization improved. We train a set of similar models to fit the data in different ways using a penalty on the alignment of their input gradients. We show theoretically and empirically that this induces the learning of more com-plex predictive patterns. OOD generalization fundamentally requires information beyond i. i.d. examples, such as multiple training environ-ments, counterfactual examples, or other side information. Our approach shows that we can defer this requirement to an independent model selection stage. We obtain SOTA re-sults in visual recognition on biased data and generalization across visual domains. The method - the first to evade the simplicity bias - highlights the need for a better under-standing and control of inductive biases in deep learning.

Neural networks trained with ERM (empirical risk minimization) sometimes learn unintended decision rules, in particular when their training data is biased, i.e., when training labels are strongly correlated with undesirable features. To prevent a network from learning such features, recent methods augment training data such that examples displaying spurious correlations (i.e., bias-aligned examples) become a minority, whereas the other, bias-conflicting examples become prevalent. However, these approaches are sometimes difficult to train and scale to real-world data because they rely on generative models or disentangled representations. We propose an alternative based on mixup, a popular augmentation that creates convex combinations of training examples. Our method, coined SelecMix, applies mixup to contradicting pairs of examples, defined as showing either (i) the same label but dissimilar biased features, or (ii) different labels but similar biased features. Identifying such pairs requires comparing examples with respect to unknown biased features. For this, we utilize an auxiliary contrastive model with the popular heuristic that biased features are learned preferentially during training. Experiments on standard benchmarks demonstrate the effectiveness of the method, in particular when label noise complicates the identification of bias-conflicting examples.

We extend the task of composed image retrieval, where an input query consists of an image and short textual description of how to modify the image. Existing methods have only been applied to non-complex images within narrow domains, such as fashion products, thereby limiting the scope of study on in-depth visual reasoning in rich image and language contexts. To address this issue, we collect the Compose Image Retrieval on Real-life images (CIRR) dataset, which consists of over 36,000 pairs of crowd-sourced, open-domain images with human-generated modifying text. To extend current methods to the open-domain, we propose CIRPLANT, a transformer based model that leverages rich pre-trained vision-and-language (V&L) knowledge for modifying visual features conditioned on natural language. Retrieval is then done by nearest neighbor lookup on the modified features. We demonstrate that with a relatively simple architecture, CIRPLANT outperforms existing methods on open-domain images, while matching state-of-the-art accuracy on the existing narrow datasets, such as fashion. Together with the release of CIRR, we believe this work will inspire further research on composed image retrieval. Our dataset, code and pre-trained models are available at https://cuberick-orion.github.io/CIRR/.

We introduce an evaluation methodology for visual question answering (VQA) to better diagnose cases of shortcut learning. These cases happen when a model exploits spurious statistical regularities to produce correct answers but does not actually deploy the desired behavior. There is a need to identify possible shortcuts in a dataset and assess their use before deploying a model in the real world. The research community in VQA has focused exclusively on question-based shortcuts, where a model might, for example, answer "What is the color of the sky" with "blue" by relying mostly on the question-conditional training prior and give little weight to visual evidence. We go a step further and consider multimodal shortcuts that involve both questions and images. We first identify potential shortcuts in the popular VQA v2 training set by mining trivial predictive rules such as co-occurrences of words and visual elements. We then introduce VQA-CounterExamples (VQACE), an evaluation protocol based on our subset of CounterExamples i.e. image-question-answer triplets where our rules lead to incorrect answers. We use this new evaluation in a large-scale study of existing approaches for VQA. We demonstrate that even state-of-the-art models perform poorly and that existing techniques to reduce biases are largely ineffective in this context. Our findings suggest that past work on question-based biases in VQA has only addressed one facet of a complex issue. The code for our method is available at https://github.com/cdancette/detect-shortcuts

Generalization beyond the training distribution is a core challenge in machine learning. The common practice of mixing and shuffling examples when training neural networks may not be optimal in this regard. We show that partitioning the data into well-chosen, non-i.i.d. subsets treated as multiple training environments can guide the learning of models with better out-of-distribution generalization. We describe a training procedure to capture the patterns that are stable across environments while discarding spurious ones. The method makes a step beyond correlation-based learning: the choice of the partitioning allows injecting information about the task that cannot be otherwise recovered from the joint distribution of the training data.We demonstrate multiple use cases with the task of visual question answering, which is notorious for dataset biases. We obtain significant improvements on VQA-CP, using environments built from prior knowledge, existing meta data, or unsupervised clustering. We also get improvements on GQA using annotations of "equivalent questions", and on multi-dataset training (VQA v2 / Visual Genome) by treating them as distinct environments.

Advances in machine learning have generated increasing enthusiasm for tasks that require high-level reasoning on top of perceptual capabilities, particularly over visual data. Such tasks include, for example, image captioning, visual question answering, and visual navigation. Their evaluation is however hindered by task-specific confounding factors and dataset biases. In parallel, the existing benchmarks for abstract reasoning are limited to synthetic stimuli (e.g. images of simple shapes) and do not capture the challenges of real-world data. We propose a new large-scale benchmark to evaluates abstract reasoning over real visual data. The test involves visual questions that require operations fundamental to many high-level vision tasks, such as comparisons of counts and logical operations on complex visual properties. The benchmark measures a method's ability to infer high-level relationships and to generalise them over image-based concepts. We provide multiple training/test splits that require controlled levels of generalization. We evaluate a range of deep learning architectures, and find that existing models, including those popular for vision-and-language tasks, are unable to solve seemingly-simple instances. Models using relational networks fare better but leave substantial room for improvement.

The ongoing success of visual question answering methods has been somwehat surprising given that, at its most general, the problem requires understanding the entire variety of both visual and language stimuli. It is particularly remarkable that this success has been achieved on the basis of comparatively small datasets, given the scale of the problem. One explanation is that this has been accomplished partly by exploiting bias in the datasets rather than developing deeper multi-modal reasoning. This fundamentally limits the generalization of the method, and thus its practical applicability. We propose a method that addresses this problem by introducing counterfactuals in the training. In doing so we leverage structural causal models for counterfactual evaluation to formulate alternatives, for instance, questions that could be asked of the same image set. We show that simulating plausible alternative training data through this process results in better generalization.

The task of vision-and-language navigation (VLN) requires an agent to follow text instructions to find its way through simulated household environments. A prominent challenge is to train an agent capable of generalising to new environments at test time, rather than one that simply memorises trajectories and visual details observed during training. We propose a new learning strategy that learns both from observations and generated counterfactual environments. We describe an effective algorithm to generate counterfactual observations on the fly for VLN, as linear combinations of existing environments. Simultaneously, we encourage the agent's actions to remain stable between original and counterfactual environments through our novel training objective-effectively removing the spurious features that otherwise bias the agent. Our experiments show that this technique provides significant improvements in generalisation on benchmarks for Room-to-Room navigation and Embodied Question Answering.

Out-of-distribution (OOD) testing is increasingly popular for evaluating a machine learning system's ability to generalize beyond the biases of a training set. OOD benchmarks are designed to present a different joint distribution of data and labels between training and test time. VQA-CP has become the standard OOD benchmark for visual question answering, but we discovered three troubling practices in its current use. First, most published methods rely on explicit knowledge of the construction of the OOD splits. They often rely on inverting'' the distribution of labels, e.g. answering mostlyyes'' when the common training answer was ``no''. Second, the OOD test set is used for model selection. Third, a model's in-domain performance is assessed after retraining it on in-domain splits (VQA v2) that exhibit a more balanced distribution of labels. These three practices defeat the objective of evaluating generalization, and put into question the value of methods specifically designed for this dataset. We show that embarrassingly-simple methods, including one that generates answers at random, surpass the state of the art on some question types. We provide short- and long-term solutions to avoid these pitfalls and realize the benefits of OOD evaluation.

One of the key limitations of traditional machine learning methods is their requirement for training data that exemplifies all the information to be learned. This is a particular problem for visual question answering methods, which may be asked questions about virtually anything. The approach we propose is a step toward overcoming this limitation by searching for the information required at test time. The resulting method dynamically utilizes data from an external source, such as a large set of questions/answers or images/captions. Concretely, we learn a set of base weights for a simple VQA model, that are specifically adapted to a given question with the information specifically retrieved for this question. The adaptation process leverages recent advances in gradient-based meta learning and contributions for efficient retrieval and cross-domain adaptation. We surpass the state-of-the-art on the VQA-CP v2 benchmark and demonstrate our approach to be intrinsically more robust to out-of-distribution test data. We demonstrate the use of external non-VQA data using the MS COCO captioning dataset to support the answering process. This approach opens a new avenue for open-domain VQA systems that interface with diverse sources of data.

Top-down visual attention mechanisms have been used extensively in image captioning and visual question answering (VQA) to enable deeper image understanding through fine-grained analysis and even multiple steps of reasoning. In this work, we propose a combined bottom-up and top-down attention mechanism that enables attention to be calculated at the level of objects and other salient image regions. This is the natural basis for attention to be considered. Within our approach, the bottom-up mechanism (based on Faster R-CNN) proposes image regions, each with an associated feature vector, while the top-down mechanism determines feature weightings. Applying this approach to image captioning, our results on the MSCOCO test server establish a new state-of-the-art for the task, achieving CIDEr / SPICE / BLEU-4 scores of 117.9, 21.5 and 36.9, respectively. Demonstrating the broad applicability of the method, applying the same approach to VQA we obtain first place in the 2017 VQA Challenge.

A robot that can carry out a natural-language instruction has been a dream since before the Jetsons cartoon series imagined a life of leisure mediated by a fleet of attentive robot helpers. It is a dream that remains stubbornly distant. However, recent advances in vision and language methods have made incredible progress in closely related areas. This is significant because a robot interpreting a natural-language navigation instruction on the basis of what it sees is carrying out a vision and language process that is similar to Visual Question Answering. Both tasks can be interpreted as visually grounded sequence-to-sequence translation problems, and many of the same methods are applicable. To enable and encourage the application of vision and language methods to the problem of interpreting visually-grounded navigation instructions, we present the Matter-port3D Simulator - a large-scale reinforcement learning environment based on real imagery [11]. Using this simulator, which can in future support a range of embodied vision and language tasks, we provide the first benchmark dataset for visually-grounded natural language navigation in real buildings - the Room-to-Room (R2R) dataset1.

Deep Learning has had a transformative impact on Computer Vision, but for all of the success there is also a significant cost. This is that the models and procedures used are so complex and intertwined that it is often impossible to distinguish the impact of the individual design and engineering choices each model embodies. This ambiguity diverts progress in the field, and leads to a situation where developing a state-of-the-art model is as much an art as a science. As a step towards addressing this problem we present a massive exploration of the effects of the myriad architectural and hyperparameter choices that must be made in generating a state-of-the-art model. The model is of particular interest because it won the 2017 Visual Question Answering Challenge. We provide a detailed analysis of the impact of each choice on model performance, in the hope that it will inform others in developing models, but also that it might set a precedent that will accelerate scientific progress in the field.

This paper proposes to improve visual question answering (VQA) with structured representations of both scene contents and questions. A key challenge in VQA is to require joint reasoning over the visual and text domains. The predominant CNN/LSTM-based approach to VQA is limited by monolithic vector representations that largely ignore structure in the scene and in the question. CNN feature vectors cannot effectively capture situations as simple as multiple object instances, and LSTMs process questions as series of words, which do not reflect the true complexity of language structure. We instead propose to build graphs over the scene objects and over the question words, and we describe a deep neural network that exploits the structure in these representations. We show that this approach achieves significant improvements over the state-of-the-art, increasing accuracy from 71.2% to 74.4% in accuracy on the abstract scenes multiple-choice benchmark, and from 34.7% to 39.1% in accuracy over pairs of balanced scenes, i.e. images with fine-grained differences and opposite yes/no answers to a same question.

This paper addresses the segmentation of videos with arbitrary motion, including dynamic textures, using novel motion features and a supervised learning approach. Dynamic textures are commonplace in natural scenes, and exhibit complex patterns of appearance and motion (e.g. water, smoke, swaying foliage). These are difficult for existing segmentation algorithms, often violate the brightness constancy assumption needed for optical flow, and have complex segment characteristics beyond uniform appearance or motion. Our solution uses custom spatiotemporal filters that capture texture and motion cues, along with a novel metric-learning framework that optimizes this representation for specific objects and scenes. This is used within a hierarchical, graph-based segmentation setting, yielding state-of-the-art results for dynamic texture segmentation. We also demonstrate the applicability of our approach to general object and motion segmentation, showing significant improvements over unsupervised segmentation and results comparable to the best task specific approaches.

We propose a novel method for performing fine-grained recognition of human hand grasp types using a single monocular image to allow computational systems to better understand human hand use. In particular, we focus on recognizing challenging grasp categories which differ only by subtle variations in finger configurations. While much of the prior work on understanding human hand grasps has been based on manual detection of grasps in video, this is the first work to automate the analysis process for fine-grained grasp classification. Instead of attempting to utilize a parametric model of the hand, we propose a hand parsing framework which leverages a data-driven learning to generate a pixel-wise segmentation of a hand into finger and palm regions. The proposed approach makes use of appearance-based cues such as finger texture and hand shape to accurately determine hand parts. We then build on the hand parsing result to compute high-level grasp features to learn a supervised fine-grained grasp classifier. To validate our approach, we introduce a grasp dataset recorded with a wearable camera, where the hand and its parts have been manually segmented with pixel-wise accuracy. Our results show that our proposed automatic hand parsing technique can improve grasp classification accuracy by over 30 percentage points over a state-of-the-art grasp recognition technique.