Publications

Embedding Signals on Graphs with Unbalanced Diffusion Earth Mover's Distance

Dennis Shung

Manik Kuchroo

In modern relational machine learning it is common to encounter large graphs that arise via interactions or similarities between observation… (voir plus)s in many domains. Further, in many cases the target entities for analysis are actually signals on such graphs. We propose to compare and organize such datasets of graph signals by using an earth mover's distance (EMD) with a geodesic cost over the underlying graph. Typically, EMD is computed by optimizing over the cost of transporting one probability distribution to another over an underlying metric space. However, this is inefficient when computing the EMD between many signals. Here, we propose an unbalanced graph EMD that efficiently embeds the unbalanced EMD on an underlying graph into an

2022-04-26

IEEE International Conference on Acoustics, Speech, and Signal Processing (publié)

doi.org

arxiv.org

Embedding Signals on Graphs with Unbalanced Diffusion Earth Mover's Distance

Dennis Shung

Manik Kuchroo

In modern relational machine learning it is common to encounter large graphs that arise via interactions or similarities between observation… (voir plus)s in many domains. Further, in many cases the target entities for analysis are actually signals on such graphs. We propose to compare and organize such datasets of graph signals by using an earth mover's distance (EMD) with a geodesic cost over the underlying graph. Typically, EMD is computed by optimizing over the cost of transporting one probability distribution to another over an underlying metric space. However, this is inefficient when computing the EMD between many signals. Here, we propose an unbalanced graph EMD that efficiently embeds the unbalanced EMD on an underlying graph into an

2022-04-26

IEEE International Conference on Acoustics, Speech and Signal Processing (publié)

doi.org

arxiv.org

Multi-tract multi-symptom relationships in pediatric concussion

Guido I. Guberman

Sonja Stojanovski

Eman Nishat

Alain Ptito

Danilo Bzdok

Anne Wheeler

Maxime Descoteaux

The heterogeneity of white matter damage and symptoms in concussion has been identified as a major obstacle to therapeutic innovation. In co… (voir plus)ntrast, most diffusion MRI (dMRI) studies on concussion have traditionally relied on group-comparison approaches that average out heterogeneity. To leverage, rather than average out, concussion heterogeneity, we combined dMRI and multivariate statistics to characterize multi-tract multi-symptom relationships. Using cross-sectional data from 306 previously concussed children aged 9–10 from the Adolescent Brain Cognitive Development Study, we built connectomes weighted by classical and emerging diffusion measures. These measures were combined into two informative indices, the first representing microstructural complexity, the second representing axonal density. We deployed pattern-learning algorithms to jointly decompose these connectivity features and 19 symptom measures. Early multi-tract multi-symptom pairs explained the most covariance and represented broad symptom categories, such as a general problems pair, or a pair representing all cognitive symptoms, and implicated more distributed networks of white matter tracts. Further pairs represented more specific symptom combinations, such as a pair representing attention problems exclusively, and were associated with more localized white matter abnormalities. Symptom representation was not systematically related to tract representation across pairs. Sleep problems were implicated across most pairs, but were related to different connections across these pairs. Expression of multi-tract features was not driven by sociodemographic and injury-related variables, as well as by clinical subgroups defined by the presence of ADHD. Analyses performed on a replication dataset showed consistent results. Using a double-multivariate approach, we identified clinically-informative, cross-demographic multi-tract multi-symptom relationships. These results suggest that rather than clear one-to-one symptom-connectivity disturbances, concussions may be characterized by subtypes of symptom/connectivity relationships. The symptom/connectivity relationships identified in multi-tract multi-symptom pairs were not apparent in single-tract/single-symptom analyses. Future studies aiming to better understand connectivity/symptom relationships should take into account multi-tract multi-symptom heterogeneity. Financial support for this work came from a Vanier Canada Graduate Scholarship from the Canadian Institutes of Health Research (G.I.G.), an Ontario Graduate Scholarship (S.S.), a Restracomp Research Fellowship provided by the Hospital for Sick Children (S.S.), an Institutional Research Chair in Neuroinformatics (M.D.), as well as a Natural Sciences and Engineering Research Council CREATE grant (M.D.).

2022-04-26

eLife (publié)

doi.org

A Probabilistic Perspective on Reinforcement Learning via Supervised Learning

Alexandre Piché

Rafael Pardinas

David Vázquez

Christopher Pal

2022-04-26

ICLR.cc/2022/Workshop/GPL (poster)

openreview.net

Chunked Autoregressive GAN for Conditional Waveform Synthesis

Max Morrison

Prem Seetharaman

Conditional waveform synthesis models learn a distribution of audio waveforms given conditioning such as text, mel-spectrograms, or MIDI. Th… (voir plus)ese systems employ deep generative models that model the waveform via either sequential (autoregressive) or parallel (non-autoregressive) sampling. Generative adversarial networks (GANs) have become a common choice for non-autoregressive waveform synthesis. However, state-of-the-art GAN-based models produce artifacts when performing mel-spectrogram inversion. In this paper, we demonstrate that these artifacts correspond with an inability for the generator to learn accurate pitch and periodicity. We show that simple pitch and periodicity conditioning is insufficient for reducing this error relative to using autoregression. We discuss the inductive bias that autoregression provides for learning the relationship between instantaneous frequency and phase, and show that this inductive bias holds even when autoregressively sampling large chunks of the waveform during each forward pass. Relative to prior state-of-the-art GAN-based models, our proposed model, Chunked Autoregressive GAN (CARGAN) reduces pitch error by 40-60%, reduces training time by 58%, maintains a fast generation speed suitable for real-time or interactive applications, and maintains or improves subjective quality.

2022-04-24

International Conference on Learning Representations (Accept (Poster))

doi.org

openreview.net

ClimateGAN: Raising Climate Change Awareness by Generating Images of Floods

Victor Schmidt

Alexandra Luccioni

Mélisande Teng

Tianyu Zhang

Alexia Reynaud

Sunand Raghupathi

Gautier Cosne

Adrien Juraver

Vahe Vardanyan

Alex Hernández-García

Yoshua Bengio

Climate change is a major threat to humanity, and the actions required to prevent its catastrophic consequences include changes in both poli… (voir plus)cy-making and individual behaviour. However, taking action requires understanding the effects of climate change, even though they may seem abstract and distant. Projecting the potential consequences of extreme climate events such as flooding in familiar places can help make the abstract impacts of climate change more concrete and encourage action. As part of a larger initiative to build a website that projects extreme climate events onto user-chosen photos, we present our solution to simulate photo-realistic floods on authentic images. To address this complex task in the absence of suitable training data, we propose ClimateGAN, a model that leverages both simulated and real data for unsupervised domain adaptation and conditional image generation. In this paper, we describe the details of our framework, thoroughly evaluate components of our architecture and demonstrate that our model is capable of robustly generating photo-realistic flooding.

2022-04-24

International Conference on Learning Representations (Accept (Poster))

doi.org

openreview.net

Compositional Attention: Disentangling Search and Retrieval

Sarthak Mittal

Sharath Chandra Raparthy

Irina Rish

Yoshua Bengio

Guillaume Lajoie

Multi-head, key-value attention is the backbone of the widely successful Transformer model and its variants. This attention mechanism uses m… (voir plus)ultiple parallel key-value attention blocks (called heads), each performing two fundamental computations: (1) search - selection of a relevant entity from a set via query-key interactions, and (2) retrieval - extraction of relevant features from the selected entity via a value matrix. Importantly, standard attention heads learn a rigid mapping between search and retrieval. In this work, we first highlight how this static nature of the pairing can potentially: (a) lead to learning of redundant parameters in certain tasks, and (b) hinder generalization. To alleviate this problem, we propose a novel attention mechanism, called Compositional Attention, that replaces the standard head structure. The proposed mechanism disentangles search and retrieval and composes them in a dynamic, flexible and context-dependent manner through an additional soft competition stage between the query-key combination and value pairing. Through a series of numerical experiments, we show that it outperforms standard multi-head attention on a variety of tasks, including some out-of-distribution settings. Through our qualitative analysis, we demonstrate that Compositional Attention leads to dynamic specialization based on the type of retrieval needed. Our proposed mechanism generalizes multi-head attention, allows independent scaling of search and retrieval, and can easily be implemented in lieu of standard attention heads in any network architecture.

2022-04-24

International Conference on Learning Representations (Accept (Spotlight))

doi.org

openreview.net

Constructing a Good Behavior Basis for Transfer Using Generalized Policy Updates

Safa Alver

Doina Precup

We study the problem of learning a good set of policies, so that when combined together, they can solve a wide variety of unseen reinforceme… (voir plus)nt learning tasks with no or very little new data. Specifically, we consider the framework of generalized policy evaluation and improvement, in which the rewards for all tasks of interest are assumed to be expressible as a linear combination of a fixed set of features. We show theoretically that, under certain assumptions, having access to a specific set of diverse policies, which we call a set of independent policies, can allow for instantaneously achieving high-level performance on all possible downstream tasks which are typically more complex than the ones on which the agent was trained. Based on this theoretical analysis, we propose a simple algorithm that iteratively constructs this set of policies. In addition to empirically validating our theoretical results, we compare our approach with recently proposed diverse policy set construction methods and show that, while others fail, our approach is able to build a behavior basis that enables instantaneous transfer to all possible downstream tasks. We also show empirically that having access to a set of independent policies can better bootstrap the learning process on downstream tasks where the new reward function cannot be described as a linear combination of the features. Finally, we demonstrate how this policy set can be useful in a lifelong reinforcement learning setting.

2022-04-24

International Conference on Learning Representations (Accept (Poster))

doi.org

openreview.net

Continuous-Time Meta-Learning with Forward Mode Differentiation

Drawing inspiration from gradient-based meta-learning methods with infinitely small gradient steps, we introduce Continuous-Time Meta-Learni… (voir plus)ng (COMLN), a meta-learning algorithm where adaptation follows the dynamics of a gradient vector field. Specifically, representations of the inputs are meta-learned such that a task-specific linear classifier is obtained as a solution of an ordinary differential equation (ODE). Treating the learning process as an ODE offers the notable advantage that the length of the trajectory is now continuous, as opposed to a fixed and discrete number of gradient steps. As a consequence, we can optimize the amount of adaptation necessary to solve a new task using stochastic gradient descent, in addition to learning the initial conditions as is standard practice in gradient-based meta-learning. Importantly, in order to compute the exact meta-gradients required for the outer-loop updates, we devise an efficient algorithm based on forward mode differentiation, whose memory requirements do not scale with the length of the learning trajectory, thus allowing longer adaptation in constant memory. We provide analytical guarantees for the stability of COMLN, we show empirically its efficiency in terms of runtime and memory usage, and we illustrate its effectiveness on a range of few-shot image classification problems.

2022-04-24

ICLR (Accept (Spotlight))

doi.org

openreview.net

Coordination Among Neural Modules Through a Shared Global Workspace

Nan Rosemary Ke

Nasim Rahaman

Jonathan Binas

Charles Blundell

Michael Mozer

Yoshua Bengio

Deep learning has seen a movement away from representing examples with a monolithic hidden state towards a richly structured state. For exam… (voir plus)ple, Transformers segment by position, and object-centric architectures decompose images into entities. In all these architectures, interactions between different elements are modeled via pairwise interactions: Transformers make use of self-attention to incorporate information from other positions; object-centric architectures make use of graph neural networks to model interactions among entities. However, pairwise interactions may not achieve global coordination or a coherent, integrated representation that can be used for downstream tasks. In cognitive science, a global workspace architecture has been proposed in which functionally specialized components share information through a common, bandwidth-limited communication channel. We explore the use of such a communication channel in the context of deep learning for modeling the structure of complex environments. The proposed method includes a shared workspace through which communication among different specialist modules takes place but due to limits on the communication bandwidth, specialist modules must compete for access. We show that capacity limitations have a rational basis in that (1) they encourage specialization and compositionality and (2) they facilitate the synchronization of otherwise independent specialists.

2022-04-24

ICLR (Accept (Oral))

doi.org

openreview.net

Exposing Query Identification for Search Transparency

Ruohan Li

Jianxiang Li

Bhaskar Mitra

Fernando Diaz

Asia J. Biega

Search systems control the exposure of ranked content to searchers. In many cases, creators value not only the exposure of their content but… (voir plus), moreover, an understanding of the specific searches where the content is surfaced. The problem of identifying which queries expose a given piece of content in the ranking results is an important and relatively under-explored search transparency challenge. Exposing queries are useful for quantifying various issues of search bias, privacy, data protection, security, and search engine optimization. Exact identification of exposing queries in a given system is computationally expensive, especially in dynamic contexts such as web search. We explore the feasibility of approximate exposing query identification (EQI) as a retrieval task by reversing the role of queries and documents in two classes of search systems: dense dual-encoder models and traditional BM25 models. We then propose how this approach can be improved through metric learning over the retrieval embedding space. We further derive an evaluation metric to measure the quality of a ranking of exposing queries, as well as conducting an empirical analysis focusing on various practical aspects of approximate EQI. Overall, our work contributes a novel conception of transparency in search systems and computational means of achieving it.

2022-04-24

ACM Web Conference (publié)

doi.org

arxiv.org

Generalized Natural Gradient Flows in Hidden Convex-Concave Games and Gans

Andjela Mladenovic

Iosif Sakos

Gauthier Gidel

Georgios Piliouras

2022-04-24

International Conference on Learning Representations (Accept (Poster))

openreview.net