Publications

Author Correction: Visualizing structure and transitions in high-dimensional biological data
Kevin R. Moon
David van Dijk
Zheng Wang
Scott Gigante
Daniel B. Burkhardt
William S. Chen
Kristina Yim
Antonia van den Elzen
Matthew Hirn
Ronald R. Coifman
Natalia Ivanova
Guy Wolf
Smita Krishnaswamy
Tess C. Branon
Justin A. Bosch
Ariana D. Sanchez
Namrata D. Udeshi
Tanya Svinkina
Steven A. Carr
Jessica L. Feldman … (see 2 more)
Norbert Perrimon
Alice Y. Ting
2019-11-19
Nature Biotechnology (published)
doi.org
Nonlinear chance-constrained problems with applications to hydro scheduling
Andrea Lodi
Enrico Malaguti
Giacomo Nannicini
Dimitri Thomopulos
2019-11-13
Mathematical programming (published)
doi.org
SST'19 - Software and Systems Traceability
Jan-Philipp Steghöfer
Nan Niu
Jin L.C. Guo
Anas Mahmoud
Traceability is the ability to relate di erent artifacts during the development and operation of a system to each other. It enables program … (see more)comprehension, change impact analysis, and facilitates the cooperation of engineers from di erent disciplines. The 10th International Workshop on Software and Systems Traceability (former International Workshop on Traceability in Emerging Forms of Software Engineering, TEFSE), explored the role and impact of traceability in modern software and systems development. The event brought together researchers and practitioners to examine the challenges of recovering, maintaining, and utilizing traceability for the myriad forms of software and systems engineering artifacts. SST'19 was a highly interactive working event focused on discussing the main problems related to software traceability in particular in the context of opportunities and challenges posed by the recent progress in Arti cial Intelligence techniques and proposing possible solutions for such problems.
2019-11-13
ACM SIGSOFT Softw. Eng. Notes (published)
doi.org
Selective Brain Damage: Measuring the Disparate Impact of Model Pruning
Sara Hooker
Aaron Courville
Yann Dauphin
Andrea Frome
Neural network pruning techniques have demonstrated it is possible to remove the majority of weights in a network with surprisingly little d… (see more)egradation to test set accuracy. However, this measure of performance conceals significant differences in how different classes and images are impacted by pruning. We find that certain examples, which we term pruning identified exemplars (PIEs), and classes are systematically more impacted by the introduction of sparsity. Removing PIE images from the test-set greatly improves top-1 accuracy for both pruned and non-pruned models. These hard-to-generalize-to images tend to be mislabelled, of lower image quality, depict multiple objects or require fine-grained classification. These findings shed light on previously unknown trade-offs, and suggest that a high degree of caution should be exercised before pruning is used in sensitive domains.
2019-11-12
arXiv.org (preprint)
openreview.net
openreview.net
What Do Compressed Deep Neural Networks Forget
Sara Hooker
Aaron Courville
Gregory Clark
Yann Dauphin
Andrea Frome
Deep neural network pruning and quantization techniques have demonstrated it is possible to achieve high levels of compression with surprisi… (see more)ngly little degradation to test set accuracy. However, this measure of performance conceals significant differences in how different classes and images are impacted by model compression techniques. We find that models with radically different numbers of weights have comparable top-line performance metrics but diverge considerably in behavior on a narrow subset of the dataset. This small subset of data points, which we term Pruning Identified Exemplars (PIEs) are systematically more impacted by the introduction of sparsity. Compression disproportionately impacts model performance on the underrepresented long-tail of the data distribution. PIEs over-index on atypical or noisy images that are far more challenging for both humans and algorithms to classify. Our work provides intuition into the role of capacity in deep neural networks and the trade-offs incurred by compression. An understanding of this disparate impact is critical given the widespread deployment of compressed models in the wild.
2019-11-12
ArXiv (preprint)
arxiv.org
arxiv.org
Defining ‘actionable’ high- costhealth care use: results using the Canadian Institute for Health Information population grouping methodology
Maureen Anderson
Crawford W. Revie
Henrik Stryhn
Cordell Neudorf
Yvonne Rosehart
Wenbin Li
Meriç Osman
David L Buckeridge
Laura C. Rosella
Walter P. Wodchis
2019-11-09
International Journal for Equity in Health (published)
doi.org
Preventing Posterior Collapse in Sequence VAEs with Pooling
Teng Long
Yanshuai Cao
Jackie CK Cheung
Variational Autoencoders (VAEs) hold great potential for modelling text, as they could in theory separate high-level semantic and syntactic … (see more)properties from local regularities of natural language. Practically, however, VAEs with autoregressive decoders often suffer from posterior collapse, a phenomenon where the model learns to ignore the latent variables, causing the sequence VAE to degenerate into a language model. Previous works attempt to solve this problem with complex architectural changes or costly optimization schemes. In this paper, we argue that posterior collapse is caused in part by the encoder network failing to capture the input variabilities. We verify this hypothesis empirically and propose a straightforward fix using pooling. This simple technique effectively prevents posterior collapse, allowing the model to achieve significantly better data log-likelihood than standard sequence VAEs. Compared to the previous SOTA on preventing posterior collapse, we are able to achieve comparable performances while being significantly faster.
2019-11-09
ArXiv (preprint)
arxiv.org
arxiv.org
Sidewalk Environment for Visual Navigation
Martin Weiß
Simon Chamorro
Roger Girgis
Margaux Luck
Samira Ebrahimi Kahou
Joseph Cohen
Derek Nowrouzezahrai
Doina Precup
Florian Golemo
Chris Pal
This dataset contains low and high resolution panoramic images, coordinates, labels and a connectivity graph. In order to run this simulated… (see more) environment, you will need at least one copy of the panoramic images, which are available in low (84x224 pixels) or high resolution (1280x3840 pixels). For more information, visit https://mweiss17.github.io/SEVN/.
2019-11-06
Polytechnique Montréal (accepted)
doi.org
Adversarial target-invariant representation learning for domain generalization
Isabela Albuquerque
Joao Monteiro
Tiago Falk
Ioannis Mitliagkas
In many applications of machine learning, the training and test set data come from different distributions, or domains. A number of domain g… (see more)eneralization strategies have been introduced with the goal of achieving good performance on out-of-distribution data. In this paper, we propose an adversarial approach to the problem. We propose a process that enforces pair-wise domain invariance while training a feature extractor over a diverse set of domains. We show that this process ensures invariance to any distribution that can be expressed as a mixture of the training domains. Following this insight, we then introduce an adversarial approach in which pair-wise divergences are estimated and minimized. Experiments on two domain generalization benchmarks for object recognition (i.e., PACS and VLCS) show that the proposed method yields higher average accuracy on the target domains in comparison to previously introduced adversarial strategies, as well as recently proposed methods based on learning invariant representations.
2019-11-02
arXiv.org (preprint)
dblp.uni-trier.de
Cascaded Gaussian Processes for Data-efficient Robot Dynamics Learning
Sahand Rezaei-Shoshtari
David Meger
Inna Sharf
Motivated by the recursive Newton-Euler formulation, we propose a novel cascaded Gaussian process learning framework for the inverse dynamic… (see more)s of robot manipulators. This approach leads to a significant dimensionality reduction which in turn results in better learning and data efficiency. We explore two formulations for the cascading: the inward and outward, both along the manipulator chain topology. The learned modeling is tested in conjunction with the classical inverse dynamics model (semi-parametric) and on its own (non-parametric) in the context of feed-forward control of the arm. Experimental results are obtained with Jaco 2 six-DOF and SARCOS seven-DOF manipulators for randomly defined sinusoidal motions of the joints in order to evaluate the performance of cascading against the standard GP learning. In addition, experiments are conducted using Jaco 2 on a task emulating a pouring maneuver. Results indicate a consistent improvement in learning speed with the inward cascaded GP model and an overall improvement in data efficiency and generalization.
2019-11-02
2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (published)
doi.org
arxiv.org
Deep Generative Modeling of LiDAR Data
Lucas Caccia
Herke van Hoof
Aaron Courville
Joelle Pineau
Building models capable of generating structured output is a key challenge for AI and robotics. While generative models have been explored o… (see more)n many types of data, little work has been done on synthesizing lidar scans, which play a key role in robot mapping and localization. In this work, we show that one can adapt deep generative models for this task by unravelling lidar scans into a 2D point map. Our approach can generate high quality samples, while simultaneously learning a meaningful latent representation of the data. We demonstrate significant improvements against state-of-the-art point cloud generation methods. Furthermore, we propose a novel data representation that augments the 2D signal with absolute positional information. We show that this helps robustness to noisy and imputed input; the learned model can recover the underlying lidar scan from seemingly uninformative data
2019-11-02
IEEE/RSJ International Conference on Intelligent Robots and Systems (published)
doi.org
arxiv.org
Generalizing to unseen domains via distribution matching
Isabela Albuquerque
Joao Monteiro
Mohammad Javad Darvishi Bayazi
Tiago Falk
Ioannis Mitliagkas
Supervised learning results typically rely on assumptions of i.i.d. data. Unfortunately, those assumptions are commonly violated in practice… (see more). In this work, we tackle this problem by focusing on domain generalization: a formalization where the data generating process at test time may yield samples from never-before-seen domains (distributions). Our work relies on a simple lemma: by minimizing a notion of discrepancy between all pairs from a set of given domains, we also minimize the discrepancy between any pairs of mixtures of domains. Using this result, we derive a generalization bound for our setting. We then show that low risk over unseen domains can be achieved by representing the data in a space where (i) the training distributions are indistinguishable, and (ii) relevant information for the task at hand is preserved. Minimizing the terms in our bound yields an adversarial formulation which estimates and minimizes pairwise discrepancies. We validate our proposed strategy on standard domain generalization benchmarks, outperforming a number of recently introduced methods. Notably, we tackle a real-world application where the underlying data corresponds to multi-channel electroencephalography time series from different subjects, each considered as a distinct domain.
2019-11-02
ArXiv (preprint)
arxiv.org
arxiv.org