Techniques enabling large language models (LLMs) to "think more" by generating and attending to intermediate reasoning steps have shown promise in solving complex problems. However, the standard approaches generate sequences of discrete tokens immediately before responding, and so they can incur significant latency costs and be challenging to optimize. In this work, we demonstrate that a frozen LLM can be augmented with an offline coprocessor that operates on the model's key-value (kv) cache. This coprocessor augments the cache with a set of latent embeddings designed to improve the fidelity of subsequent decoding. We train this coprocessor using the language modeling loss from the decoder on standard pretraining data, while keeping the decoder itself frozen. This approach enables the model to learn, in an end-to-end differentiable fashion, how to distill additional computation into its kv-cache. Because the decoder remains unchanged, the coprocessor can operate offline and asynchronously, and the language model can function normally if the coprocessor is unavailable or if a given cache is deemed not to require extra computation. We show experimentally that when a cache is augmented, the decoder achieves lower perplexity on numerous subsequent tokens. Furthermore, even without any task-specific training, our experiments demonstrate that cache augmentation consistently improves performance across a range of reasoning-intensive tasks.

As we scale to more massive machine learning models, the frequent synchronization demands inherent in data-parallel approaches create significant slowdowns, posing a critical challenge to further scaling. Recent work develops an approach (DiLoCo) that relaxes synchronization demands without compromising model quality. However, these works do not carefully analyze how DiLoCo's behavior changes with model size. In this work, we study the scaling law behavior of DiLoCo when training LLMs under a fixed compute budget. We focus on how algorithmic factors, including number of model replicas, hyperparameters, and token budget affect training in ways that can be accurately predicted via scaling laws. We find that DiLoCo scales both predictably and robustly with model size. When well-tuned, DiLoCo scales better than data-parallel training with model size, and can outperform data-parallel training even at small model sizes. Our results showcase a more general set of benefits of DiLoCo than previously documented, including increased optimal batch sizes, improved downstream generalization with scale, and improved evaluation loss for a fixed token budget.

Recent progress in using machine learning models for reasoning tasks has been driven by novel model architectures, large-scale pre-training protocols, and dedicated reasoning datasets for fine-tuning. In this work, to further pursue these advances, we introduce a new data generator for machine reasoning that integrates with an embodied agent. The generated data consists of templated text queries and answers, matched with world-states encoded into a database. The world-states are a result of both world dynamics and the actions of the agent. We show the results of several baseline models on instantiations of train sets. These include pre-trained language models fine-tuned on a text-formatted representation of the database, and graph-structured Transformers operating on a knowledge-graph representation of the database. We find that these models can answer some questions about the world-state, but struggle with others. These results hint at new research directions in designing neural reasoning models and database representations. Code to generate the data and train the models will be released at github.com/facebookresearch/neuralmemory

Large language models have been shown to struggle with multi-step reasoning, and do not retain previous reasoning steps for future use. We propose a simple method for solving both of these problems by allowing the model to take Self-Notes. Unlike recent chain-of-thought or scratchpad approaches, the model can deviate from the input context at any time to explicitly think and write down its thoughts. This allows the model to perform reasoning on the fly as it reads the context and even integrate previous reasoning steps, thus enhancing its memory with useful information and enabling multi-step reasoning. Experiments across a wide variety of tasks demonstrate that our method can outperform chain-of-thought and scratchpad methods by taking Self-Notes that interleave the input text.

Despite recent improvements in open-domain dialogue models, state-of-the-art models are trained and evaluated on short conversations with little context.In contrast, the long-term conversation setting has hardly been studied.In this work we collect and release a humanhuman dataset consisting of multiple chat sessions whereby the speaking partners learn about each other's interests and discuss the things they have learnt from past sessions.We show how existing models trained on existing datasets perform poorly in this long-term conversation setting in both automatic and human evaluations, and we study long-context models that can perform much better.In particular, we find retrieval-augmented methods and methods with an ability to summarize and recall previous conversations outperform the standard encoder-decoder architectures currently considered state-of-the-art.* We use this term colloquially, see Agranoff et al. (1965) for evidence of goldfish long-term memory.

Abstract While improving neural dialogue agents’ factual accuracy is the object of much research, another important aspect of communication, less studied in the setting of neural dialogue, is transparency about ignorance. In this work, we analyze to what extent state-of-the-art chit-chat models are linguistically calibrated in the sense that their verbalized expression of doubt (or confidence) matches the likelihood that the model’s responses are factually incorrect (or correct). We find that these models are poorly calibrated, yet we show that likelihood of correctness can accurately be predicted. By incorporating such metacognitive features into the training of a controllable generation model, we obtain a dialogue agent with greatly improved linguistic calibration.

Attention mechanisms have shown promising results in sequence modeling tasks that require long-term memory. Recent work investigated mechanisms to reduce the computational cost of preserving and storing memories. However, not all content in the past is equally important to remember. We propose Expire-Span, a method that learns to retain the most important information and expire the irrelevant information. This forgetting of memories enables Transformers to scale to attend over tens of thousands of previous timesteps efficiently, as not all states from previous timesteps are preserved. We demonstrate that Expire-Span can help models identify and retain critical information and show it can achieve strong performance on reinforcement learning tasks specifically designed to challenge this functionality. Next, we show that Expire-Span can scale to memories that are tens of thousands in size, setting a new state of the art on incredibly long context tasks such as character-level language modeling and a frame-by-frame moving objects task. Finally, we analyze the efficiency of Expire-Span compared to existing approaches and demonstrate that it trains faster and uses less memory.

Humans can learn and reason under substantial uncertainty in a space of infinitely many compositional, productive concepts. For example, if a scene with two blue spheres qualifies as “daxy,” one can reason that the underlying concept may require scenes to have “only blue spheres” or “only spheres” or “only two objects.” In contrast, standard benchmarks for compositional reasoning do not explicitly capture a notion of reasoning under uncertainty or evaluate compositional concept acquisition. We introduce a new benchmark, Compositional Reasoning Under Uncertainty (CURI) that instantiates a series of few-shot, meta-learning tasks in a productive concept space to evaluate different aspects of systematic generalization under uncertainty, including splits that test abstract understandings of disentangling, productive generalization, learning boolean operations, variable binding, etc. Importantly, we also contribute a model-independent “compositionality gap” to evaluate the difficulty of generalizing out-of-distribution along each of these axes, allowing objective comparison of the difficulty of each compositional split. Evaluations across a range of modeling choices and splits reveal substantial room for improvement on the proposed benchmark.

In recent years, there have been significant advances in building end-to-end Machine Learning (ML) systems that learn at scale. But most of these systems are: (a) isolated (perception, speech, or language only); (b) trained on static datasets. On the other hand, in the field of robotics, large-scale learning has always been difficult. Supervision is hard to gather and real world physical interactions are expensive.In this work we introduce and open-source droidlet, a modular, heterogeneous agent architecture and platform. It allows us to exploit both large-scale static datasets in perception and language and sophisticated heuristics often used in robotics; and provides tools for interactive annotation. Furthermore, it brings together perception, language and action onto one platform, providing a path towards agents that learn from the richness of real world interactions.

Prithviraj Ammanabrolu, Jack Urbanek, Margaret Li, Arthur Szlam, Tim Rocktäschel, Jason Weston. Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2021.

We investigate the training of sparse layers that use different parameters for different inputs based on hashing in large Transformer models. Specifically, we modify the feedforward layer to hash to different sets of weights depending on the current token, over all tokens in the sequence. We show that this procedure either outperforms or is competitive with learning-to-route mixture-of-expert methods such as Switch Transformers and BASE Layers, while requiring no routing parameters or extra terms in the objective function such as a load balancing loss, and no sophisticated assignment algorithm. We study the performance of different hashing techniques, hash sizes and input features, and show that balanced and random hashes focused on the most local features work best, compared to either learning clusters or using longer-range context. We show our approach works well both on large language modeling and dialogue tasks, and on downstream fine-tuning tasks.

Current large-scale auto-regressive language models display impressive fluency and can generate convincing text. In this work we start by asking the question: Can the generations of these models be reliably distinguished from real text by statistical discriminators? We find experimentally that the answer is affirmative when we have access to the training data for the model, and guardedly affirmative even if we do not. This suggests that the auto-regressive models can be improved by incorporating the (globally normalized) discriminators into the generative process. We give a formalism for this using the Energy-Based Model framework, and show that it indeed improves the results of the generative models, measured both in terms of perplexity and in terms of human evaluation.

Procedurally generating cohesive and interesting game environments is challenging and time-consuming. In order for the relationships between the game elements to be natural, common-sense has to be encoded into arrangement of the elements. In this work, we investigate a machine learning approach for world creation using content from the multi-player text adventure game environment LIGHT (Urbanek et al. 2019). We introduce neural network based models to compositionally arrange locations, characters, and objects into a coherent whole. In addition to creating worlds based on existing elements, our models can generate new game content. Humans can also leverage our models to interactively aid in worldbuilding. We show that the game environments created with our approach are cohesive, diverse, and preferred by human evaluators compared to other machine learning based world construction algorithms.

We propose a semantic parsing dataset focused on instruction-driven communication with an agent in the game Minecraft 1 .The dataset consists of 7K human utterances and their corresponding parses.Given proper world state, the parses can be interpreted and executed in game.We report the performance of baseline models, and analyze their successes and failures.

Standard RL algorithms assume fixed environment dynamics and require a significant amount of interaction to adapt to new environments. We introduce Policy-Dynamics Value Functions (PD-VF), a novel approach for rapidly adapting to dynamics different from those previously seen in training. PD-VF explicitly estimates the cumulative reward in a space of policies and environments. An ensemble of conventional RL policies is used to gather experience on training environments, from which embeddings of both policies and environments can be learned. Then, a value function conditioned on both embeddings is trained. At test time, a few actions are sufficient to infer the environment embedding, enabling a policy to be selected by maximizing the learned value function (which requires no additional environment interaction). We show that our method can rapidly adapt to new dynamics on a set of MuJoCo domains.

Consistency is a long standing issue faced by dialogue models. In this paper, we frame the consistency of dialogue agents as natural language inference (NLI) and create a new natural language inference dataset called Dialogue NLI. We propose a method which demonstrates that a model trained on Dialogue NLI can be used to improve the consistency of a dialogue model, and evaluate the method with human evaluation and with automatic metrics on a suite of evaluation sets designed to measure a dialogue model’s consistency.

Jack Urbanek, Angela Fan, Siddharth Karamcheti, Saachi Jain, Samuel Humeau, Emily Dinan, Tim Rocktäschel, Douwe Kiela, Arthur Szlam, Jason Weston. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). 2019.

Research on 2D and 3D generative models typically focuses on the final artifact being created, e.g., an image or a 3D structure. Unlike 2D image generation, the generation of 3D objects in the real world is commonly constrained by the process and order in which the object is constructed. For instance, gravity needs to be taken into account when building a block tower. In this paper, we explore the prediction of ordered actions to construct 3D objects. Instead of predicting actions based on physical constraints, we propose learning through observing human actions. To enable large-scale data collection, we use the Minecraft1 environment. We introduce 3D-Craft, a new dataset of 2,500 Minecraft houses each built by human players sequentially from scratch. To learn from these human action sequences, we propose an order-aware 3D generative model called VoxelCNN. In contrast to other 3D generative models which either have no explicit order (e.g. holistic generation with 3DGAN [35]), or follow a simple heuristic order (e.g. raster-scan), VoxelCNN is trained to imitate human building order with spatial awareness. We also transferred the order to other dataset such as ShapeNet[10]. The 3D-Craft dataset, models, and benchmark system will be made publicly available, which may inspire new directions for future research exploration. https://github.com/facebookresearch/VoxelCNN.

Saizheng Zhang, Emily Dinan, Jack Urbanek, Arthur Szlam, Douwe Kiela, Jason Weston. Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2018.

This paper considers the problem of inferring image labels from images when only a few annotated examples are available at training time. This setup is often referred to as low-shot learning, where a standard approach is to retrain the last few layers of a convolutional neural network learned on separate classes for which training examples are abundant. We consider a semi-supervised setting based on a large collection of images to support label propagation. This is possible by leveraging the recent advances on large-scale similarity graph construction. We show that despite its conceptual simplicity, scaling label propagation up to hundred millions of images leads to state of the art accuracy in the low-shot learning regime.

Generative Adversarial Networks (GANs) have achieved remarkable results in the task of generating realistic natural images. In most successful applications, GAN models share two common aspects: solving a challenging saddle point optimization problem, interpreted as an adversarial game between a generator and a discriminator functions; and parameterizing the generator and the discriminator as deep convolutional neural networks. The goal of this paper is to disentangle the contribution of these two factors to the success of GANs. In particular, we introduce Generative Latent Optimization (GLO), a framework to train deep convolutional generators using simple reconstruction losses. Throughout a variety of experiments, we show that GLO enjoys many of the desirable properties of GANs: synthesizing visually-appealing samples, interpolating meaningfully between samples, and performing linear arithmetic with noise vectors; all of this without the adversarial optimization scheme.

We consider the multi-agent reinforcement learning setting with imperfect information. The reward function depends on the hidden goals of both agents, so the agents must infer the other players’ goals from their observed behavior in order to maximize their returns. We propose a new approach for learning in these domains: Self Other-Modeling (SOM), in which an agent uses its own policy to predict the other agent’s actions and update its belief of their hidden goal in an online manner. We evaluate this approach on three different tasks and show that the agents are able to learn better policies using their estimate of the other players’ goals, in both cooperative and competitive settings.

The tasks that an agent will need to solve often are not known during training. However, if the agent knows which properties of the environment are important then, after learning how its actions affect those properties, it may be able to use this knowledge to solve complex tasks without training specifically for them. Towards this end, we consider a setup in which an environment is augmented with a set of user defined attributes that parameterize the features of interest. We propose a method that learns a policy for transitioning between “nearby” sets of attributes, and maintains a graph of possible transitions. Given a task at test time that can be expressed in terms of a target set of attributes, and a current state, our model infers the attributes of the current state and searches over paths through attribute space to get a high level plan, and then uses its low level policy to execute the plan. We show in 3D block stacking, grid-world games, and StarCraft that our model is able to generalize to longer, more complex tasks at test time by composing simpler learned policies.

Training convolutional networks (CNNs) that fit on a single GPU with minibatch stochastic gradient descent has become effective in practice. However, there is still no effective method for training large networks that do not fit in the memory of a few GPU cards, or for parallelizing CNN training. In this work we show that a simple hard mixture of experts model can be efficiently trained to good effect on large scale hashtag (multilabel) prediction tasks. Mixture of experts models are not new [7, 3], but in the past, researchers have had to devise sophisticated methods to deal with data fragmentation. We show empirically that modern weakly supervised data sets are large enough to support naive partitioning schemes where each data point is assigned to a single expert. Because the experts are independent, training them in parallel is easy, and evaluation is cheap for the size of the model. Furthermore, we show that we can use a single decoding layer for all the experts, allowing a unified feature embedding space. We demonstrate that it is feasible (and in fact relatively painless) to train far larger models than could be practically trained with standard CNN architectures, and that the extra capacity can be well used on current datasets.

Recent years have witnessed intense development of randomized methods for low-rank approximation. These methods target principal component analysis and the calculation of truncated singular value decompositions. The present article presents an essentially black-box, foolproof implementation for Mathworks' MATLAB, a popular software platform for numerical computation. As illustrated via several tests, the randomized algorithms for low-rank approximation outperform or at least match the classical deterministic techniques (such as Lanczos iterations run to convergence) in basically all respects: accuracy, computational efficiency (both speed and memory usage), ease-of-use, parallelizability, and reliability. However, the classical procedures remain the methods of choice for estimating spectral norms and are far superior for calculating the least singular values and corresponding singular vectors (or singular subspaces).

Although RNNs have been shown to be power- ful tools for processing sequential data, finding architectures or optimization strategies that al- low them to model very long term dependencies is still an active area of research. In this work, we carefully analyze two synthetic datasets orig- inally outlined in (Hochreiter & Schmidhuber, 1997) which are used to evaluate the ability of RNNs to store information over many time steps. We explicitly construct RNN solutions to these problems, and using these constructions, illumi- nate both the problems themselves and the way in which RNNs store different types of information in their hidden states. These constructions fur- thermore explain the success of recent methods that specify unitary initializations or constraints on the transition matrices.

We introduce a theoretical and algorithmic framework for multi-way graph partitioning that relies on a multiplicative cut-based objective. We refer to this objective as the Product Cut. We provide a detailed investigation of the mathematical properties of this objective and an effective algorithm for its optimization. The proposed model has strong mathematical underpinnings, and the corresponding algorithm achieves state-of-the-art performance on benchmark data sets.

Many tasks in AI require the collaboration of multiple agents. Typically, the communication protocol between agents is manually specified and not altered during training. In this paper we explore a simple neural model, called CommNet, that uses continuous communication for fully cooperative tasks. The model consists of multiple agents and the communication between them is learned alongside their policy. We apply this model to a diverse set of tasks, demonstrating the ability of the agents to learn to communicate amongst themselves, yielding improved performance over non-communicative agents and baselines. In some cases, it is possible to interpret the language devised by the agents, revealing simple but effective strategies for solving the task at hand.

A complex-valued convolutional network (convnet) implements the repeated application of the following composition of three operations, recursively applying the composition to an input vector of nonnegative real numbers: (1) convolution with complex-valued vectors, followed by (2) taking the absolute value of every entry of the resulting vectors, followed by (3) local averaging. For processing real-valued random vectors, complex-valued convnets can be viewed as data-driven multiscale windowed power spectra, data-driven multiscale windowed absolute spectra, data-driven multiwavelet absolute values, or (in their most general configuration) data-driven nonlinear multiwavelet packets. Indeed, complex-valued convnets can calculate multiscale windowed spectra when the convnet filters are windowed complex-valued exponentials. Standard real-valued convnets, using rectified linear units (ReLUs), sigmoidal (e.g., logistic or tanh) nonlinearities, or max pooling, for example, do not obviously exhibit the same exact correspondence with data-driven wavelets (whereas for complex-valued convnets, the correspondence is much more than just a vague analogy). Courtesy of the exact correspondence, the remarkably rich and rigorous body of mathematical analysis for wavelets applies directly to (complex-valued) convnets.

In this paper we introduce a generative model capable of producing high quality samples of natural images. Our approach uses a cascade of convolutional networks (convnets) within a Laplacian pyramid framework to generate images in a coarse-to-fine fashion. At each level of the pyramid a separate generative convnet model is trained using the Generative Adversarial Nets (GAN) approach. Samples drawn from our model are of significantly higher quality than existing models. In a quantitive assessment by human evaluators our CIFAR10 samples were mistaken for real images around 40% of the time, compared to 10% for GAN samples. We also show samples from more diverse datasets such as STL10 and LSUN.

We introduce a neural network with a recurrent attention model over a possibly large external memory. The architecture is a form of Memory Network (Weston et al., 2015) but unlike the model in that work, it is trained end-to-end, and hence requires significantly less supervision during training, making it more generally applicable in realistic settings. It can also be seen as an extension of RNNsearch to the case where multiple computational steps (hops) are performed per output symbol. The flexibility of the model allows us to apply it to tasks as diverse as (synthetic) question answering and to language modeling. For the former our approach is competitive with Memory Networks, but with less supervision. For the latter, on the Penn TreeBank and Text8 datasets our approach demonstrates comparable performance to RNNs and LSTMs. In both cases we show that the key concept of multiple computational hops yields improved results.

Feature tracking in video is a crucial task in computer vision. Usually, the tracking problem is handled one feature at a time, using a single-feature tracker like the Kanade-Lucas-Tomasi algorithm, or one of its derivatives. While this approach works quite well when dealing with high-quality video and "strong" features, it often falters when faced with dark and noisy video containing low-quality features. We present a framework for jointly tracking a set of features, which enables sharing information between the different features in the scene. We show that our method can be employed to track features for both rigid and non-rigid motions (possibly of few moving bodies) even when some features are occluded. Furthermore, it can be used to significantly improve tracking results in poorly-lit scenes (where there is a mix of good and bad features). Our approach does not require direct modeling of the structure or the motion of the scene, and runs in real time on a single CPU core.

We describe a method for learning a piecewise affine approximation to a mapping f : ℝd→ Rpgiven a labeled training set of examples {x1,..., xn} = X ⊂ ℝdand targets {y1= f(x1), ..., yn= f(xn)} = Y ⊂ ℝp. The method first trains a binary subdivision tree that splits across hyperplanes in X corresponding to high variance directions in Y. A fixed number K of affine regressors of rank q are then trained via a K-means like iterative algorithm, where each leaf must vote on its best fit mapping, and each mapping is updated as the best fit for the collection of leaves that chose it.

Pooling operators construct non-linear representations by cascading a redundant linear transform, followed by a point-wise nonlinearity and a local aggregation, typically implemented with a \ell_p norm. Their efficiency in recognition architectures is based on their ability to locally contract the input space, but also on their capacity to retain as much stable information as possible. We address this latter question by computing the upper and lower Lipschitz bounds of \ell_p pooling operators for p=1, 2, ∞as well as their half-rectified equivalents, which give sufficient conditions for the design of invertible pooling layers. Numerical experiments on MNIST and image patches confirm that pooling layers can be inverted with phase recovery algorithms. Moreover, the regularity of the inverse pooling, controlled by the lower Lipschitz constant, is empirically verified with a nearest neighbor regression.

In this paper, we propose a new unsupervised feature learning framework, namely Deep Sparse Coding (DeepSC), that extends sparse coding to a multi-layer architecture for visual object recognition tasks. The main innovation of the framework is that it connects the sparse-encoders from different layers by a sparse-to-dense module. The sparse-to-dense module is a composition of a local spatial pooling step and a low-dimensional embedding process, which takes advantage of the spatial smoothness information in the image. As a result, the new method is able to learn multiple layers of sparse representations of the image which capture features at a variety of abstraction levels and simultaneously preserve the spatial smoothness between the neighboring image patches. Combining the feature representations from multiple layers, DeepSC achieves the state-of-the-art performance on multiple object recognition tasks.

We study the related problems of denoising images corrupted by impulsive noise and blind inpainting (i.e., inpainting when the deteriorated region is unknown). Our basic approach is to model the set of patches of pixels in an image as a union of low-dimensional subspaces, corrupted by sparse but perhaps large magnitude noise. For this purpose, we develop a robust and iterative method for single subspace modeling and extend it to an iterative algorithm for modeling multiple subspaces. We prove convergence for both algorithms and carefully compare our methods with other recent ideas for such robust modeling. We demonstrate state-of-the-art performance of our method for both imaging problems.

It has recently been shown that only a small number of samples from a low-rank matrix are necessary to reconstruct the entire matrix. We bring this to bear on computer vision problems that utilize low-dimensional subspaces, demonstrating that subsampling can improve computation speed while still allowing for accurate subspace learning. We present GRASTA, Grassmannian Robust Adaptive Subspace Tracking Algorithm, an online algorithm for robust subspace estimation from randomly subsampled data. We consider the specific application of background and foreground separation in video, and we assess GRASTA on separation accuracy and computation time. In one benchmark video example [16], GRASTA achieves a separation rate of 46.3 frames per second, even when run in MATLAB on a personal laptop.

We present a simple and fast geometric method for modeling data by a union of affine subspaces. The method begins by forming a collection of local best-fit affine subspaces, i.e., subspaces approximating the data in local neighborhoods. The correct sizes of the local neighborhoods are determined automatically by the Jones’ β 2 numbers (we prove under certain geometric conditions that our method finds the optimal local neighborhoods). The collection of subspaces is further processed by a greedy selection procedure or a spectral method to generate the final model. We discuss applications to tracking-based motion segmentation and clustering of faces under different illuminating conditions. We give extensive experimental evidence demonstrating the state of the art accuracy and speed of the suggested algorithms on these problems and also on synthetic hybrid linear data as well as the MNIST handwritten digits data; and we demonstrate how to use our algorithms for fast determination of the number of affine subspaces.

This work describes a conceptually simple method for structured sparse coding and dictionary design. Supposing a dictionary with K atoms, we introduce a structure as a set of penalties or interactions between every pair of atoms. We describe modifications of standard sparse coding algorithms for inference in this setting, and describe experiments showing that these algorithms are efficient. We show that interesting dictionaries can be learned for interactions that encode tree structures or locally connected structures. Finally, we show that our framework allows us to learn the values of the interactions from the data, rather than having them pre-specified.

The hybrid linear modeling problem is to identify a set of d-dimensional affine sets in RD. It arises, for example, in object tracking and structure from motion. The hybrid linear model can be considered as the second simplest (behind linear) manifold model of data. In this paper we will present a very simple geometric method for hybrid linear modeling based on selecting a set of local best fit flats that minimize a global ℓ1error measure. The size of the local neighborhoods is determined automatically by the Jones' β2numbers; it is proven under certain geometric conditions that good local neighborhoods exist and are found by our method. We also demonstrate how to use this algorithm for fast determination of the number of affine subspaces. We give extensive experimental evidence demonstrating the state of the art accuracy and speed of the algorithm on synthetic and real hybrid linear data.

We will describe an alternating direction (aka split Bregman) method for solving problems of the form minu∥Au - f∥2+ η∥u∥1such that u ≥ 0, where A is an m×n matrix, and η is a nonnegative parameter. The algorithm works especially well for solving large numbers of small to medium overdetermined problems (i.e. m > n) with a fixed A. We will demonstrate applications in the analysis of hyperspectral images.

The k q-flats algorithm is a generalization of the popular k-means algorithm where q dimensional best fit affine sets replace centroids as the cluster prototypes. In this work, a modification of the k q-flats framework for pattern classification is introduced. The basic idea is to replace the original reconstruction only energy, which is optimized to obtain the k affine spaces, by a new energy that incorporates discriminative terms. This way, the actual classification task is introduced as part of the design and optimization. The presentation of the proposed framework is complemented with experimental results, showing that the method is computationally very efficient and gives excellent results on standard supervised learning benchmarks.