Publications

Iterated Learning for Emergent Systematicity in VQA
Ankit Vani
Max Schwarzer
Yuchen Lu
Eeshan Dhekane
Aaron Courville
Although neural module networks have an architectural bias towards compositionality, they require gold standard layouts to generalize system… (voir plus)atically in practice. When instead learning layouts and modules jointly, compositionality does not arise automatically and an explicit pressure is necessary for the emergence of layouts exhibiting the right structure. We propose to address this problem using iterated learning, a cognitive science theory of the emergence of compositional languages in nature that has primarily been applied to simple referential games in machine learning. Considering the layouts of module networks as samples from an emergent language, we use iterated learning to encourage the development of structure within this language. We show that the resulting layouts support systematic generalization in neural agents solving the more complex task of visual question-answering. Our regularized iterated learning method can outperform baselines without iterated learning on SHAPES-SyGeT (SHAPES Systematic Generalization Test), a new split of the SHAPES dataset we introduce to evaluate systematic generalization, and on CLOSURE, an extension of CLEVR also designed to test systematic generalization. We demonstrate superior performance in recovering ground-truth compositional program structure with limited supervision on both SHAPES-SyGeT and CLEVR.
2021-01-11
ICLR.cc/2021/Conference (présentation orale)
doi.org
openreview.net
Learning to Live with Dale's Principle: ANNs with Separate Excitatory and Inhibitory Units
Jonathan Cornford
Damjan Kalajdzievski
Marco Leite
Amélie Lamarquette
Dimitri M. Kullmann
Blake Richards
A bstract The units in artificial neural networks (ANNs) can be thought of as abstraction… (voir plus)s of biological neurons, and ANNs are increasingly used in neuroscience research. However, there are many important differences between ANN units and real neurons. One of the most notable is the absence of Dale’s principle, which ensures that biological neurons are either exclusively excitatory or inhibitory. Dale’s principle is typically left out of ANNs because its inclusion impairs learning. This is problematic, because one of the great advantages of ANNs for neuroscience research is their ability to learn complicated, realistic tasks. Here, by taking inspiration from feedforward inhibitory interneurons in the brain we show that we can develop ANNs with separate populations of excitatory and inhibitory units that learn just as well as standard ANNs. We call these networks Dale’s ANNs (DANNs). We present two insights that enable DANNs to learn well: (1) DANNs are related to normalization schemes, and can be initialized such that the inhibition centres and standardizes the excitatory activity, (2) updates to inhibitory neuron parameters should be scaled using corrections based on the Fisher Information matrix. These results demonstrate how ANNs that respect Dale’s principle can be built without sacrificing learning performance, which is important for future work using ANNs as models of the brain. The results may also have interesting implications for how inhibitory plasticity in the real brain operates.
2021-01-11
ICLR.cc/2021/Conference (poster)
doi.org
openreview.net
Recurrent Independent Mechanisms
Anirudh Goyal
Alex Lamb
Jordan Hoffmann
Shagun Sodhani
Sergey Levine
Yoshua Bengio
Bernhard Schölkopf
Learning modular structures which reflect the dynamics of the environment can lead to better generalization and robustness to changes which … (voir plus)only affect a few of the underlying causes. We propose Recurrent Independent Mechanisms (RIMs), a new recurrent architecture in which multiple groups of recurrent cells operate with nearly independent transition dynamics, communicate only sparingly through the bottleneck of attention, and are only updated at time steps where they are most relevant. We show that this leads to specialization amongst the RIMs, which in turn allows for dramatically improved generalization on tasks where some factors of variation differ systematically between training and evaluation.
2021-01-11
ICLR.cc/2021/Conference (spotlight)
doi.org
openreview.net
RNNLogic: Learning Logic Rules for Reasoning on Knowledge Graphs
Meng Qu
Junkun Chen
Louis-Pascal Xhonneux
Yoshua Bengio
Jian Tang
This paper studies learning logic rules for reasoning on knowledge graphs. Logic rules provide interpretable explanations when used for pred… (voir plus)iction as well as being able to generalize to other tasks, and hence are critical to learn. Existing methods either suffer from the problem of searching in a large search space (e.g., neural logic programming) or ineffective optimization due to sparse rewards (e.g., techniques based on reinforcement learning). To address these limitations, this paper proposes a probabilistic model called RNNLogic. RNNLogic treats logic rules as a latent variable, and simultaneously trains a rule generator as well as a reasoning predictor with logic rules. We develop an EM-based algorithm for optimization. In each iteration, the reasoning predictor is first updated to explore some generated logic rules for reasoning. Then in the E-step, we select a set of high-quality rules from all generated rules with both the rule generator and reasoning predictor via posterior inference; and in the M-step, the rule generator is updated with the rules selected in the E-step. Experiments on four datasets prove the effectiveness of RNNLogic.
2021-01-11
ICLR.cc/2021/Conference (poster)
doi.org
openreview.net
Spatially Structured Recurrent Modules
Nasim Rahaman
Anirudh Goyal
Muhammad Waleed Gondal
Manuel Wuthrich
Stefan Bauer
Yash Sharma
Yoshua Bengio
Bernhard Schölkopf
Capturing the structure of a data-generating process by means of appropriate inductive biases can help in learning models that generalise we… (voir plus)ll and are robust to changes in the input distribution. While methods that harness spatial and temporal structures find broad application, recent work has demonstrated the potential of models that leverage sparse and modular structure using an ensemble of sparingly interacting modules. In this work, we take a step towards dynamic models that are capable of simultaneously exploiting both modular and spatiotemporal structures. To this end, we model the dynamical system as a collection of autonomous but sparsely interacting sub-systems that interact according to a learned topology which is informed by the spatial structure of the underlying system. This gives rise to a class of models that are well suited for capturing the dynamics of systems that only offer local views into their state, along with corresponding spatial locations of those views. On the tasks of video prediction from cropped frames and multi-agent world modelling from partial observations in the challenging Starcraft2 domain, we find our models to be more robust to the number of available views and better capable of generalisation to novel tasks without additional training than strong baselines that perform equally well or better on the training distribution.
2021-01-11
ICLR.cc/2021/Conference (poster)
openreview.net
openreview.net
Attention Based Pruning for Shift Networks
Ghouthi Boukli hacene
Carlos Lassance
Vincent Gripon
Matthieu Courbariaux
Yoshua Bengio
In many application domains such as computer vision, Convolutional Layers (CLs) are key to the accuracy of deep learning methods. However, i… (voir plus)t is often required to assemble a large number of CLs, each containing thousands of parameters, in order to reach state-of-the-art accuracy, thus resulting in complex and demanding systems that are poorly fitted to resource-limited devices. Recently, methods have been proposed to replace the generic convolution operator by the combination of a shift operation and a simpler 1x1 convolution. The resulting block, called Shift Layer (SL), is an efficient alternative to CLs in the sense it allows to reach similar accuracies on various tasks with faster computations and fewer parameters. In this contribution, we introduce Shift Attention Layers (SALs), which extend SLs by using an attention mechanism that learns which shifts are the best at the same time the network function is trained. We demonstrate SALs are able to outperform vanilla SLs (and CLs) on various object recognition benchmarks while significantly reducing the number of float operations and parameters for the inference.
2021-01-09
2020 25th International Conference on Pattern Recognition (ICPR) (publié)
doi.org
arxiv.org
The patient advisor, an organizational resource as a lever for an enhanced oncology patient experience (PAROLE-onco): a longitudinal multiple case study protocol
M. P. Pomey
M. de Guise
M. Desforges
K. Bouchard
C. Vialaron
L. Normandin
M. Iliescu-Nelea
I. Fortin
I. Ganache
C. Régis
Z. Rosberger
D. Charpentier
L. Bélanger
M. Dorval
D. P. Ghadiri
M. Lavoie-Tremblay
A. Boivin
J. F. Pelletier
N. Fernandez
A. M. Danino
Quebec is one of the Canadian provinces with the highest rates of cancer incidence and prevalence. A study by the Rossy Cancer Network (RCN)… (voir plus) of McGill university assessed six aspects of the patient experience among cancer patients and found that emotional support is the aspect most lacking. To improve this support, trained patient advisors (PAs) can be included as full-fledged members of the healthcare team, given that PA can rely on their knowledge with experiencing the disease and from using health and social care services to accompany cancer patients, they could help to round out the health and social care services offer in oncology. However, the feasibility of integrating PAs in clinical oncology teams has not been studied. In this multisite study, we will explore how to integrate PAs in clinical oncology teams and, under what conditions this can be successfully done. We aim to better understand effects of this PA intervention on patients, on the PAs themselves, the health and social care team, the administrators, and on the organization of services and to identify associated ethical and legal issues. We will conduct six mixed methods longitudinal case studies. Qualitative data will be used to study the integration of the PAs into clinical oncology teams and to identify the factors that are facilitators and inhibitors of the process, the associated ethical and legal issues, and the challenges that the PAs experience. Quantitative data will be used to assess effects on patients, PAs and team members, if any, of the PA intervention. The results will be used to support oncology programs in the integration of PAs into their healthcare teams and to design a future randomized pragmatic trial to evaluate the impact of PAs as full-fledged members of clinical oncology teams on cancer patients’ experience of emotional support throughout their care trajectory. This study will be the first to integrate PAs as full-fledged members of the clinical oncology team and to assess possible clinical and organizational level effects. Given the unique role of PAs, this study will complement the body of research on peer support and patient navigation. An additional innovative aspect of this study will be consideration of the ethical and legal issues at stake and how to address them in the health care organizations.
2021-01-03
BMC Health Services Research (publié)
doi.org
Seeing Through your Skin: Recognizing Objects with a Novel Visuotactile Sensor
Francois Hogan
M. Jenkin
Sahand Rezaei-Shoshtari
Yogesh Girdhar
David Meger
Gregory Dudek
We introduce a new class of vision-based sensor and associated algorithmic processes that combine visual imaging with high-resolution tactil… (voir plus)e sending, all in a uniform hardware and computational architecture. We demonstrate the sensor’s efficacy for both multi-modal object recognition and metrology. Object recognition is typically formulated as an unimodal task, but by combining two sensor modalities we show that we can achieve several significant performance improvements. This sensor, named the See-Through-your-Skin sensor (STS), is designed to provide rich multi-modal sensing of contact surfaces. Inspired by recent developments in optical tactile sensing technology, we address a key missing feature of these sensors: the ability to capture a visual perspective of the region beyond the contact surface. Whereas optical tactile sensors are typically opaque, we present a sensor with a semitransparent skin that has the dual capabilities of acting as a tactile sensor and/or as a visual camera depending on its internal lighting conditions. This paper details the design of the sensor, showcases its dual sensing capabilities, and presents a deep learning architecture that fuses vision and touch. We validate the ability of the sensor to classify household objects, recognize fine textures, and infer their physical properties both through numerical simulations and experiments with a smart countertop prototype.
2021-01-02
2021 IEEE Winter Conference on Applications of Computer Vision (WACV) (publié)
doi.org
arxiv.org
Abg-CoQA: Clarifying Ambiguity in Conversational Question Answering
Meiqi Guo
Mingda Zhang
Siva Reddy
Malihe Alikhani
Effective communication is about the dissemination of properly worded meaningful ideas/messages that are comprehensible to both sen… (voir plus)der and receiver and which ultimately can attract the desired response or feedback. For machines to engage in a conversation, it is therefore essential to enable them to clarify ambiguity and achieve a common ground. We introduce Abg-CoQA, a novel dataset for clarifying ambiguity in Conversational Question Answering systems. Our dataset contains 9k questions with answers where 1k questions are ambiguous, obtained from 4k text passages from five diverse domains. For ambiguous questions, a clarification conversational turn is collected. We evaluate strong language generation models and conversational question answering models on Abg-CoQA. The best-performing system achieves a BLEU-1 score of 12.9% on generating clarification question, which is 27.9 points behind human performance (40.8%); and a F1 score of 40.1% on question answering after clarification, which is 35.1 points behind human performance (75.2%), indicating there is ample room for improvement.
2020-12-31
Conference on Automated Knowledge Base Construction (publié)
doi.org
openreview.net
Accounting for Variance in Machine Learning Benchmarks
Xavier Bouthillier
Pierre Delaunay
Mirko Bronzi
Assya Trofimov
Brennan Nichyporuk
Justin Szeto
Naz Sepah
Edward Raff
Kanika Madan
Vikram Voleti
Samira Ebrahimi Kahou
Vincent Michalski
Dmitriy Serdyuk
Tal Arbel
Chris Pal
Gael Varoquaux
Pascal Vincent
Strong empirical evidence that one machine-learning algorithm A outperforms another one B ideally calls for multiple trials optimizing the l… (voir plus)earning pipeline over sources of variation such as data sampling, data augmentation, parameter initialization, and hyperparameters choices. This is prohibitively expensive, and corners are cut to reach conclusions. We model the whole benchmarking process, revealing that variance due to data sampling, parameter initialization and hyperparameter choice impact markedly the results. We analyze the predominant comparison methods used today in the light of this variance. We show a counter-intuitive result that adding more sources of variation to an imperfect estimator approaches better the ideal estimator at a 51 times reduction in compute cost. Building on these results, we study the error rate of detecting improvements, on five different deep-learning tasks/architectures. This study leads us to propose recommendations for performance comparisons.
2020-12-31
MLSys (publié)
doi.org
arxiv.org
ADEPT: An Adjective-Dependent Plausibility Task
Ali Emami
Ian Porada
A.R. Olteanu
Kaheer Suleman
Adam Trischler
Jackie CK Cheung
2020-12-31
Annual Meeting of the Association for Computational Linguistics (publié)
doi.org
Adversarial Feature Desensitization
Pouya Bashivan
Reza Bayat
Adam Ibrahim
Kartik Ahuja
Mojtaba Faramarzi
Touraj Laleh
Blake Richards
Irina Rish
Neural networks are known to be vulnerable to adversarial attacks -- slight but carefully constructed perturbations of the inputs which can … (voir plus)drastically impair the network's performance. Many defense methods have been proposed for improving robustness of deep networks by training them on adversarially perturbed inputs. However, these models often remain vulnerable to new types of attacks not seen during training, and even to slightly stronger versions of previously seen attacks. In this work, we propose a novel approach to adversarial robustness, which builds upon the insights from the domain adaptation field. Our method, called Adversarial Feature Desensitization (AFD), aims at learning features that are invariant towards adversarial perturbations of the inputs. This is achieved through a game where we learn features that are both predictive and robust (insensitive to adversarial attacks), i.e. cannot be used to discriminate between natural and adversarial data. Empirical results on several benchmarks demonstrate the effectiveness of the proposed approach against a wide range of attack types and attack strengths. Our code is available at https://github.com/BashivanLab/afd.
2020-12-31
Advances in Neural Information Processing Systems 34 (NeurIPS 2021) (publié)
doi.org
openreview.net