Publications

Techniques for Symbol Grounding with SATNet
Sever Topan
David Rolnick
Xujie Si
Many experts argue that the future of artificial intelligence is limited by the field's ability to integrate symbolic logical reasoning into… (see more) deep learning architectures. The recently proposed differentiable MAXSAT solver, SATNet, was a breakthrough in its capacity to integrate with a traditional neural network and solve visual reasoning problems. For instance, it can learn the rules of Sudoku purely from image examples. Despite its success, SATNet was shown to succumb to a key challenge in neurosymbolic systems known as the Symbol Grounding Problem: the inability to map visual inputs to symbolic variables without explicit supervision ("label leakage"). In this work, we present a self-supervised pre-training pipeline that enables SATNet to overcome this limitation, thus broadening the class of problems that SATNet architectures can solve to include datasets where no intermediary labels are available at all. We demonstrate that our method allows SATNet to attain full accuracy even with a harder problem setup that prevents any label leakage. We additionally introduce a proofreading method that further improves the performance of SATNet architectures, beating the state-of-the-art on Visual Sudoku.
2021-12-06
Conference on Neural Information Processing Systems (Accept (Spotlight))
doi.org
openreview.net
The Causal-Neural Connection: Expressiveness, Learnability, and Inference
Kevin Xia
Kai-Zhan Lee
Yoshua Bengio
Elias Bareinboim
One of the central elements of any causal inference is an object called structural causal model (SCM), which represents a collection of mech… (see more)anisms and exogenous sources of random variation of the system under investigation (Pearl, 2000). An important property of many kinds of neural networks is universal approximability: the ability to approximate any function to arbitrary precision. Given this property, one may be tempted to surmise that a collection of neural nets is capable of learning any SCM by training on data generated by that SCM. In this paper, we show this is not the case by disentangling the notions of expressivity and learnability. Specifically, we show that the causal hierarchy theorem (Thm. 1, Bareinboim et al., 2020), which describes the limits of what can be learned from data, still holds for neural models. For instance, an arbitrarily complex and expressive neural net is unable to predict the effects of interventions given observational data alone. Given this result, we introduce a special type of SCM called a neural causal model (NCM), and formalize a new type of inductive bias to encode structural constraints necessary for performing causal inferences. Building on this new class of models, we focus on solving two canonical tasks found in the literature known as causal identification and estimation. Leveraging the neural toolbox, we develop an algorithm that is both sufficient and necessary to determine whether a causal effect can be learned from data (i.e., causal identifiability); it then estimates the effect whenever identifiability holds (causal estimation). Simulations corroborate the proposed approach.
2021-12-06
Conference on Neural Information Processing Systems (Accept (Poster))
doi.org
openreview.net
The functional specialization of visual cortex emerges from training parallel pathways with self-supervised predictive learning
Shahab Bakhtiari
Patrick Mineault
Tim Lillicrap
Christopher C. Pack
Blake A. Richards
The visual system of mammals is comprised of parallel, hierarchical specialized pathways. Different pathways are specialized in so far as th… (see more)ey use representations that are more suitable for supporting specific downstream behaviours. In particular, the clearest example is the specialization of the ventral (“what”) and dorsal (“where”) pathways of the visual cortex. These two pathways support behaviours related to visual recognition and movement, respectively. To-date, deep neural networks have mostly been used as models of the ventral, recognition pathway. However, it is unknown whether both pathways can be modelled with a single deep ANN. Here, we ask whether a single model with a single loss function can capture the properties of both the ventral and the dorsal pathways. We explore this question using data from mice, who like other mammals, have specialized pathways that appear to support recognition and movement behaviours. We show that when we train a deep neural network architecture with two parallel pathways using a self-supervised predictive loss function, we can outperform other models in fitting mouse visual cortex. Moreover, we can model both the dorsal and ventral pathways. These results demonstrate that a self-supervised predictive learning approach applied to parallel pathway architectures can account for some of the functional specialization seen in mammalian visual systems.
2021-12-06
NEURIPS (Accept (Spotlight))
doi.org
openreview.net
Your head is there to move you around: Goal-driven models of the primate dorsal pathway
Patrick J Mineault
Shahab Bakhtiari
Blake A Richards
Christopher C Pack
Neurons in the dorsal visual pathway of the mammalian brain are selective for motion stimuli, with the complexity of stimulus representation… (see more)s increasing along the hierarchy. This progression is similar to that of the ventral visual pathway, which is well characterized by artificial neural networks (ANNs) optimized for object recognition. In contrast, there are no image-computable models of the dorsal stream with comparable explanatory power. We hypothesized that the properties of dorsal stream neurons could be explained by a simple learning objective: the need for an organism to orient itself during self-motion. To test this hypothesis, we trained a 3D ResNet to predict an agent’s self-motion parameters from visual stimuli in a simulated environment. We found that the responses in this network accounted well for the selectivity of neurons in a large database of single-neuron recordings from the dorsal visual stream of non-human primates. In contrast, ANNs trained on an action recognition dataset through supervised or self-supervised learning could not explain responses in the dorsal stream, despite also being trained on naturalistic videos with moving objects. These results demonstrate that an ecologically relevant cost function can account for dorsal stream properties in the primate brain.
2021-12-06
NEURIPS (Accept (Spotlight))
doi.org
openreview.net
Patient health records and whole viral genomes from an early SARS-CoV-2 outbreak in a Quebec hospital reveal features associated with favorable outcomes
Bastien Paré
Marieke Rozendaal
Sacha Morin
Raphaël Poujol
Fatima Mostefai
Shawn M. Simpson
Jean-Christophe Grenier
Léa Kaufmann
Henry Xing
Miguelle Sanchez
Ariane Yechouron
Ronald Racette
Julie G. Hussin
Guy Wolf
Ivan Pavlov
Martin A. Smith
The first confirmed case of COVID-19 in Quebec, Canada, occurred at Verdun Hospital on February 25, 2020. A month later, a localized outbrea… (see more)k was observed at this hospital. We performed tiled amplicon whole genome nanopore sequencing on nasopharyngeal swabs from all SARS-CoV-2 positive samples from 31 March to 17 April 2020 in 2 local hospitals to assess the viral diversity of the outbreak. We report 264 viral genomes from 242 individuals (both staff and patients) with associated clinical features and outcomes, as well as longitudinal samples, technical replicates and the first publicly disseminated SARS-CoV-2 genomes in Quebec. Viral lineage assessment identified multiple subclades in both hospitals, with a predominant subclade in the Verdun outbreak, indicative of hospital-acquired transmission. Dimensionality reduction identified two subclades that evaded supervised lineage assignment methods, including Pangolin, and identified certain symptoms (headache, myalgia and sore throat) that are significantly associated with favorable patient outcomes. We also address certain limitations of standard SARS-CoV-2 bioinformatics procedures, notably when presented with multiple viral haplotypes.
2021-12-01
PLOS ONE (published)
doi.org
Cognitive health mediates the effect of hippocampal volume on COVID‐19‒related knowledge or anxiety change during the COVID‐19 pandemic
Min Su Kang
Julie Ottoy
Stijn Servaes
Firoza Z Lussier
Gleb Bezgin
Mira Chamoun
Jenna Stevenson
Danilo Bzdok
Suzanne King
Serge Gauthier
Pedro Rosa‐Neto
Our finding highlights the poorer knowledge of COVID19 and related risks in individuals with cognitive/memory impairments; the CDRSOB, indic… (see more)ative of cognitive health, significantly mediated the effect of hippocampal volume on the rate of change in anxiety or knowledge on COVID19 in our cohort. This study urges for a more effective strategy and policy about informing and educating the individual with cognitive/memory impairment on COVID19 and related risks.
2021-11-30
Alzheimer's & Dementia (published)
doi.org
Online Partisan Polarization of COVID-19
Zachary Yang
Anne Imouza
Kellin Pelrine
Sacha Lévy
Jiewen Liu
Gabrielle Desrosiers-Brisebois
Jean-François Godbout
Andre Blais
Reihaneh Rabbany
In today’s age of (mis)information, many people utilize various social media platforms in an attempt to shape public opinion on sever… (see more)al important issues, including elections and the COVID-19 pandemic. These two topics have recently become intertwined given the importance of complying with public health measures related to COVID-19 and politicians’ management of the pandemic. Motivated by this, we study the partisan polarization of COVID-19 discussions on social media. We propose and utilize a novel measure of partisan polarization to analyze more than 380 million posts from Twitter and Parler around the 2020 US presidential election. We find strong correlation between peaks in polarization and polarizing events, such as the January 6th Capitol Hill riot. We further classify each post into key COVID-19 issues of lockdown, masks, vaccines, as well as miscellaneous, to investigate both the volume and polarization on these topics and how they vary through time. Parler includes more negative discussions around lockdown and masks, as expected, but not much around vaccines. We also observe more balanced discussions on Twitter and a general disconnect between the discussions on Parler and Twitter.
2021-11-30
2021 International Conference on Data Mining Workshops (ICDMW) (unknown)
doi.org
Systematic Generalization with Edge Transformers
Leon Bergen
Timothy J. O'Donnell
Dzmitry Bahdanau
Recent research suggests that systematic generalization in natural language understanding remains a challenge for state-of-the-art neural mo… (see more)dels such as Transformers and Graph Neural Networks. To tackle this challenge, we propose Edge Transformer, a new model that combines inspiration from Transformers and rule-based symbolic AI. The first key idea in Edge Transformers is to associate vector states with every edge, that is, with every pair of input nodes -- as opposed to just every node, as it is done in the Transformer model. The second major innovation is a triangular attention mechanism that updates edge representations in a way that is inspired by unification from logic programming. We evaluate Edge Transformer on compositional generalization benchmarks in relational reasoning, semantic parsing, and dependency parsing. In all three settings, the Edge Transformer outperforms Relation-aware, Universal and classical Transformer baselines.
2021-11-30
arXiv (Cornell University) (unknown)
doi.org
arxiv.org
Tau‐load in the lingual gyrus impacts anxiety levels during the COVID‐19 pandemic in participants of longitudinal observational studies in aging
Stijn Servaes
Firoza Z Lussier
Gleb Bezgin
Yi‐Ting Wang
Jenna Stevenson
Cécile Tissot
Guillaume Elgbeili
Jaime Fernandez Arias
Joseph Therriault
Andréa Lessa Benedet
Mira Chamoun
Tharick A. Pascoal
Danilo Bzdok
Suzanne King
Serge Gauthier
Pedro Rosa‐Neto
By obtaining a better grasp on the impact of the COVID‐19 pandemic on individuals with cognitive impairment, this knowledge could be used … (see more)to improve the delivery of information to this particular group. We aimed to assess the relationship between tau deposition and the change in anxiety levels, before and during the pandemic. We hypothesized that since the pandemic, higher tau loads would lower the change in anxiety. Furthermore, we expected these anxiety levels not to be associated with COVID‐19 related stress in participants with cognitive decline. 63 participants of the Translational Biomarker of Aging and Dementia (TRIAD) cohort (cognitively healthy, N=38; cognitively impaired, N=25, of which 7 had dementia due to Alzheimer’s disease), were assessed to evaluate their individual change in anxiety levels (GAD‐7). This was done at three different timepoints, of which the latest fell during the COVID‐19 lockdown period. Two rates of change, one before and one during the pandemic, were determined using the following definition: (next timepoint – current timepoint)/time difference. In addition, at the latest timepoint, subjective stress due to COVID‐19 was measured using the Montreal Assessment of Stress related to COVID‐19 (MASC). To assess the levels of tau, standard uptake value ratios (SUVR) from previously obtained [18F]MK‐6240 PET‐scans were used. [18F]MK‐6240 tracer binding in the lingual gyrus was negatively associated with the rate of change in GAD‐7 scores after correcting for age, sex, years of education and the presence of APOE ε4, but only in cognitively impaired individuals during the pandemic (fig 1A). In addition, the GAD‐7 score at the latest timepoint was associated with stress related to COVID‐19, but only in cognitively healthy individuals (fig 1B and 1C). The presence of tau in the lingual gyrus negatively affected the rate of change in GAD‐7 scores during the COVID‐19 pandemic in individuals with cognitive impairment. This could indicate that information pertaining to the pandemic does not reach these individuals in an efficient manner. The missing association between COVID‐19 induced stress and the latest GAD‐7 scores in these individuals is a further indication of this.
2021-11-30
Alzheimer's & Dementia (unknown)
doi.org
Hypo- and hyper- sensory processing heterogeneity in Autism Spectrum Disorder
Aline Lefebvre
Julian Tillmann
Freddy Cliquet
Frederique Amsellem
Anna Maruani
Claire Leblond
Anita Beggiato
David Germanaud
Anouck Amestoy
Myriam Ly‐Le Moal
Daniel Umbricht
Christopher Chattam
Lorraine Murtagh
Manuel Bouvard
Marion Leboyer
Tony Charman
Thomas Bourgeron
Richard Delorme
Guillaume Dumas
Background. Sensory processing atypicalities are part of the core symptoms of autism spectrum disorder (ASD) and could result from an excita… (see more)tion/inhibition imbalance. Yet, the convergence level of phenotypic sensory processing atypicalities with genetic alterations in GABA-ergic and glutamatergic pathways remains poorly understood. This study aimed to characterize the distribution of hypo/hyper-sensory profile among individuals with ASD and investigate the role of deleterious mutations in GABAergic and glutamatergic pathways related genes in sensory processing atypicalities. Method. From the Short Sensory Profile (SSP) questionnaire, we defined and explored a score – the differential Short Sensory Profile (dSSP) - as a normalized and centralized hypo/hypersensitivity ratio for 1136 participants (533 with ASD, 210 first-degree relatives, and 267 controls) from two independent study samples (PARIS and LEAP). We also performed an unsupervised item-based clustering analysis on SSP items scores to validate this new categorization in terms of hypo and hyper sensitivity. We then explored the link between the dSSP score and the burden of deleterious mutations in a subset of individuals for which whole-genome sequencing data were available. Results. We observed a mean dSSP score difference between ASD and controls, driven mostly by a high dSSP score variability among groups (PARIS: p0.0001, η2 = 0.0001, LEAP: p0.0001, Cohen’s d=3.67). First-degree relatives were with an intermediate distribution variability prof
2021-11-29
(published)
doi.org
Fixing Bias in Reconstruction-Based Anomaly Detection with Lipschitz Discriminators
Alexander Tong
Guy Wolf
Smita Krishnaswamy
Anomaly detection is of great interest in fields where abnormalities need to be identified and corrected (e.g., medicine and finance). Deep … (see more)learning methods for this task often rely on autoencoder reconstruction error, sometimes in conjunction with other errors. We show that this approach exhibits intrinsic biases that lead to undesirable results. Reconstruction-based methods are sensitive to training-data outliers and simple-to-reconstruct points. Instead, we introduce a new unsupervised Lipschitz anomaly discriminator that does not suffer from these biases. Our anomaly discriminator is trained, similar to the ones used in GANs, to detect the difference between the training data and corruptions of the training data. We show that this procedure successfully detects unseen anomalies with guarantees on those that have a certain Wasserstein distance from the data or corrupted training set. These additions allow us to show improved performance on MNIST, CIFAR10, and health record data.
2021-11-27
Journal of Signal Processing Systems (unknown)
doi.org
arxiv.org
Loneliness is linked to specific subregional alterations in hippocampus-default network co-variation
Chris Zajner
R. Nathan Spreng
Danilo Bzdok
Social interaction complexity makes humans unique. But in times of social deprivation this strength risks to expose important vulnerabilitie… (see more)s. Human social neuroscience studies have placed a premium on the default network (DN). In contrast, hippocampus (HC) subfields have been intensely studied in rodents and monkeys. To bridge these two literatures, we here quantified how DN subregions systematically co-vary with specific HC subfields in the context of subjective social isolation (i.e., loneliness). By co-decomposition using structural brain scans of ∼40,000 UK Biobank participants, loneliness was specially linked to midline subregions in the uncovered DN patterns. These association cortex signatures coincided with concomitant HC patterns implicating especially CA1 and molecular layer. These patterns also showed a strong affiliation with the fornix white-matter tract and the nucleus accumbens. In addition, separable signatures of structural HC-DN co-variation had distinct associations with the genetic predisposition for loneliness at the population level.
2021-11-23
Journal of Neurophysiology (published)
doi.org