Publications

Interactive Machine Comprehension with Information Seeking Agents
Xingdi Yuan
Jie Fu
Marc-Alexandre Côté
Yi Tay
Christopher Pal
Adam Trischler
Existing machine reading comprehension (MRC) models do not scale effectively to real-world applications like web-level information retrieval… (voir plus) and question answering (QA). We argue that this stems from the nature of MRC datasets: most of these are static environments wherein the supporting documents and all necessary information are fully observed. In this paper, we propose a simple method that reframes existing MRC datasets as interactive, partially observable environments. Specifically, we “occlude” the majority of a document’s text and add context-sensitive commands that reveal “glimpses” of the hidden text to a model. We repurpose SQuAD and NewsQA as an initial case study, and then show how the interactive corpora can be used to train a model that seeks relevant information through sequential decision making. We believe that this setting can contribute in scaling models to web-level QA scenarios.
2020-06-30
Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (publié)
doi.org
arxiv.org
Words Aren’t Enough, Their Order Matters: On the Robustness of Grounding Visual Referring Expressions
Arjun Reddy Akula
Spandana Gella
Yaser Al-Onaizan
Song-Chun Zhu
Siva Reddy
Visual referring expression recognition is a challenging task that requires natural language understanding in the context of an image. We cr… (voir plus)itically examine RefCOCOg, a standard benchmark for this task, using a human study and show that 83.7% of test instances do not require reasoning on linguistic structure, i.e., words are enough to identify the target object, the word order doesn’t matter. To measure the true progress of existing models, we split the test set into two sets, one which requires reasoning on linguistic structure and the other which doesn’t. Additionally, we create an out-of-distribution dataset Ref-Adv by asking crowdworkers to perturb in-domain examples such that the target object changes. Using these datasets, we empirically show that existing methods fail to exploit linguistic structure and are 12% to 23% lower in performance than the established progress for this task. We also propose two methods, one based on contrastive learning and the other based on multi-task learning, to increase the robustness of ViLBERT, the current state-of-the-art model for this task. Our datasets are publicly available at https://github.com/aws/aws-refcocog-adv.
2020-06-30
Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (publié)
doi.org
arxiv.org
Would you Rather? A New Benchmark for Learning Machine Alignment with Cultural Values and Social Preferences
Yi Tay
Donovan Ong
Jie Fu
Alvin Chan
Nancy Chen
Anh Tuan Luu
Christopher Pal
2020-06-30
Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (publié)
doi.org
Bringing proportional recovery into proportion: Bayesian modelling of post-stroke motor impairment
Anna K Bonkhoff
Thomas Hope
Danilo Bzdok
Adrian G Guggisberg
Rachel L Hawe
Sean P Dukelow
Anne K Rehme
Gereon R Fink
Christian Grefkes
Howard Bowman
Accurate predictions of motor impairment after stroke are of cardinal importance for the patient, clinician, and healthcare system. More tha… (voir plus)n 10 years ago, the proportional recovery rule was introduced by promising that high-fidelity predictions of recovery following stroke were based only on the initially lost motor function, at least for a specific fraction of patients. However, emerging evidence suggests that this recovery rule is subject to various confounds and may apply less universally than previously assumed. Here, we systematically revisited stroke outcome predictions by applying strategies to avoid confounds and fitting hierarchical Bayesian models. We jointly analysed 385 post-stroke trajectories from six separate studies—one of the largest overall datasets of upper limb motor recovery. We addressed confounding ceiling effects by introducing a subset approach and ensured correct model estimation through synthetic data simulations. Subsequently, we used model comparisons to assess the underlying nature of recovery within our empirical recovery data. The first model comparison, relying on the conventional fraction of patients called ‘fitters’, pointed to a combination of proportional to lost function and constant recovery. ‘Proportional to lost’ here describes the original notion of proportionality, indicating greater recovery in case of a more severe initial impairment. This combination explained only 32% of the variance in recovery, which is in stark contrast to previous reports of >80%. When instead analysing the complete spectrum of subjects, ‘fitters’ and ‘non-fitters’, a combination of proportional to spared function and constant recovery was favoured, implying a more significant improvement in case of more preserved function. Explained variance was at 53%. Therefore, our quantitative findings suggest that motor recovery post-stroke may exhibit some characteristics of proportionality. However, the variance explained was substantially reduced compared to what has previously been reported. This finding motivates future research moving beyond solely behaviour scores to explain stroke recovery and establish robust and discriminating single-subject predictions.
2020-06-28
Brain : a journal of neurology (publié)
doi.org
Medical Imaging with Deep Learning: MIDL 2020 -- Short Paper Track
Tal Arbel
Ismail Ben Ayed
Marleen de Bruijne
Maxime Descoteaux
Hervé Lombaert
Chris Pal
This compendium gathers all the accepted extended abstracts from the Third International Conference on Medical Imaging with Deep Learning (M… (voir plus)IDL 2020), held in Montreal, Canada, 6-9 July 2020. Note that only accepted extended abstracts are listed here, the Proceedings of the MIDL 2020 Full Paper Track are published in the Proceedings of Machine Learning Research (PMLR).
2020-06-28
arXiv (Cornell University) (prépublication)
doi.org
arxiv.org
Object Files and Schemata: Factorizing Declarative and Procedural Knowledge in Dynamical Systems
Anirudh Goyal
Alex Lamb
Phanideep Gampa
Philippe Beaudoin
Sergey Levine
Charles Blundell
Yoshua Bengio
Michael Mozer
2020-06-28
arXiv (prépublication)
doi.org
arxiv.org
Shared and unique brain network features predict cognition, personality and mental health in childhood
Jianzhong Chen
Angela Tam
Valeria Kebets
Csaba Orbán
Leon Qi Rong Ooi
Scott Marek
Nico U.F. Dosenbach
Simon B. Eickhoff
Danilo Bzdok
Avram J. Holmes
B.T. Thomas Yeo
The manner through which individual differences in brain network organization track population-level behavioral variability is a fundamental… (voir plus) question in systems neuroscience. Recent work suggests that resting-state and task-state functional connectivity can predict specific traits at the individual level. However, the focus of most studies on single behavioral traits has come at the expense of capturing broader relationships across behaviors. Here, we utilized a large-scale dataset of 1858 typically developing children to estimate whole-brain functional network organization that is predictive of individual differences in cognition, impulsivity-related personality, and mental health during rest and task states. Predictive network features were distinct across the broad behavioral domains: cognition, personality and mental health. On the other hand, traits within each behavioral domain were predicted by highly similar network features. This is surprising given decades of research emphasizing that distinct brain networks support different mental processes. Although tasks are known to modulate the functional connectome, we found that predictive network features were similar between resting and task states. Overall, our findings reveal shared brain network features that account for individual variation within broad domains of behavior in childhood, yet are unique to different behavioral domains.
2020-06-23
bioRxiv (prépublication)
doi.org
Image-to-image Mapping with Many Domains by Sparse Attribute Transfer
Matthew Amodio
Rim Assouel
Victor Schmidt
Tristan Sylvain
Smita Krishnaswamy
Yoshua Bengio
2020-06-22
arXiv (prépublication)
doi.org
arxiv.org
Rethinking Distributional Matching Based Domain Adaptation
Bin Li
Yezhen Wang
Tong Che
Shanghang Zhang
Sicheng Zhao
Pengfei Xu
Wenzhen Zhou
Yoshua Bengio
Kurt W. Keutzer
Domain adaptation (DA) is a technique that transfers predictive models trained on a labeled source domain to an unlabeled target domain, wit… (voir plus)h the core difficulty of resolving distributional shift between domains. Currently, most popular DA algorithms are based on distributional matching (DM). However in practice, realistic domain shifts (RDS) may violate their basic assumptions and as a result these methods will fail. In this paper, in order to devise robust DA algorithms, we first systematically analyze the limitations of DM based methods, and then build new benchmarks with more realistic domain shifts to evaluate the well-accepted DM methods. We further propose InstaPBM, a novel Instance-based Predictive Behavior Matching method for robust DA. Extensive experiments on both conventional and RDS benchmarks demonstrate both the limitations of DM methods and the efficacy of InstaPBM: Compared with the best baselines, InstaPBM improves the classification accuracy respectively by
2020-06-22
ArXiv (prépublication)
arxiv.org
arxiv.org
Advantages of biologically-inspired adaptive neural activation in RNNs during learning
Victor Geadah
Giancarlo Kerg
Stefan Horoi
Guy Wolf
Guillaume Lajoie
Dynamic adaptation in single-neuron response plays a fundamental role in neural coding in biological neural networks. Yet, most neural activ… (voir plus)ation functions used in artificial networks are fixed and mostly considered as an inconsequential architecture choice. In this paper, we investigate nonlinear activation function adaptation over the large time scale of learning, and outline its impact on sequential processing in recurrent neural networks. We introduce a novel parametric family of nonlinear activation functions, inspired by input-frequency response curves of biological neurons, which allows interpolation between well-known activation functions such as ReLU and sigmoid. Using simple numerical experiments and tools from dynamical systems and information theory, we study the role of neural activation features in learning dynamics. We find that activation adaptation provides distinct task-specific solutions and in some cases, improves both learning speed and performance. Importantly, we find that optimal activation features emerging from our parametric family are considerably different from typical functions used in the literature, suggesting that exploiting the gap between these usual configurations can help learning. Finally, we outline situations where neural activation adaptation alone may help mitigate changes in input statistics in a given task, suggesting mechanisms for transfer learning optimization.
2020-06-21
ArXiv (prépublication)
doi.org
arxiv.org
HNHN: Hypergraph Networks with Hyperedge Neurons
Yihe Dong
W. Sawin
Yoshua Bengio
2020-06-21
ArXiv (prépublication)
arxiv.org
arxiv.org
Learning to Prove from Synthetic Theorems
Eser Aygün
Zafarali Ahmed
Ankit Anand
Vlad Firoiu
Xavier Glorot
Laurent Orseau
Doina Precup
Shibl Mourad
A major challenge in applying machine learning to automated theorem proving is the scarcity of training data, which is a key ingredient in t… (voir plus)raining successful deep learning models. To tackle this problem, we propose an approach that relies on training with synthetic theorems, generated from a set of axioms. We show that such theorems can be used to train an automated prover and that the learned prover transfers successfully to human-generated theorems. We demonstrate that a prover trained exclusively on synthetic theorems can solve a substantial fraction of problems in TPTP, a benchmark dataset that is used to compare state-of-the-art heuristic provers. Our approach outperforms a model trained on human-generated problems in most axiom sets, thereby showing the promise of using synthetic data for this task.
2020-06-18
ArXiv (prépublication)
arxiv.org
arxiv.org