Publications

Probability Distillation: A Caveat and Alternatives
Chin-wei Huang
Faruk Ahmed
Kundan Kumar
Alexandre Lacoste
Aaron Courville
Due to Van den Oord et al. (2018), probability distillation has recently been of interest to deep learning practitioners, where, as a practi… (voir plus)cal workaround for deploying autoregressive models in real-time applications, a student network is used to obtain quality samples in parallel. We identify a pathological optimization issue with the adopted stochastic minimization of the reverse-KL divergence: the curse of dimensionality results in a skewed gradient distribution that renders training inefficient. This means that KL-based “evaluative” training can be susceptible to poor exploration if the target distribution is highly structured. We then explore alternative principles for distillation, including one with an “instructive” signal, and show that it is possible to achieve qualitatively better results than with KL minimization.
2020-08-05
Uncertainty in Artificial Intelligence (publié)
proceedings.mlr.press
proceedings.mlr.press
Randomized Value Functions via Multiplicative Normalizing Flows
Ahmed Touati
Harsh Satija
Joshua Romoff
Joelle Pineau
Pascal Vincent
Randomized value functions offer a promising approach towards the challenge of efficient exploration in complex environments with high dimen… (voir plus)sional state and action spaces. Unlike traditional point estimate methods, randomized value functions maintain a posterior distribution over action-space values. This prevents the agent's behavior policy from prematurely exploiting early estimates and falling into local optima. In this work, we leverage recent advances in variational Bayesian neural networks and combine these with traditional Deep Q-Networks (DQN) and Deep Deterministic Policy Gradient (DDPG) to achieve randomized value functions for high-dimensional domains. In particular, we augment DQN and DDPG with multiplicative normalizing flows in order to track a rich approximate posterior distribution over the parameters of the value function. This allows the agent to perform approximate Thompson sampling in a computationally efficient manner via stochastic gradient methods. We demonstrate the benefits of our approach through an empirical comparison in high dimensional environments.
2020-08-05
Conference on Uncertainty in Artificial Intelligence (publié)
doi.org
proceedings.mlr.press
Distinct miRNA Profile of Cellular and Extracellular Vesicles Released from Chicken Tracheal Cells Following Avian Influenza Virus Infection
Kelsey O’Dowd
Mehdi Emam
Mohamed Reda El Khili
Amin Emad
Eveline M. Ibeagha-Awemu
Carl A. Gagnon
Neda Barjesteh
Innate responses provide the first line of defense against viral infections, including the influenza virus at mucosal surfaces. Communicatio… (voir plus)n and interaction between different host cells at the early stage of viral infections determine the quality and magnitude of immune responses against the invading virus. The release of membrane-encapsulated extracellular vesicles (EVs), from host cells, is defined as a refined system of cell-to-cell communication. EVs contain a diverse array of biomolecules, including microRNAs (miRNAs). We hypothesized that the activation of the tracheal cells with different stimuli impacts the cellular and EV miRNA profiles. Chicken tracheal rings were stimulated with polyI:C and LPS from Escherichia coli 026:B6 or infected with low pathogenic avian influenza virus H4N6. Subsequently, miRNAs were isolated from chicken tracheal cells or from EVs released from chicken tracheal cells. Differentially expressed (DE) miRNAs were identified in treated groups when compared to the control group. Our results demonstrated that there were 67 up-regulated miRNAs, 157 down-regulated miRNAs across all cellular and EV samples. In the next step, several genes or pathways targeted by DE miRNAs were predicted. Overall, this study presented a global miRNA expression profile in chicken tracheas in response to avian influenza viruses (AIV) and toll-like receptor (TLR) ligands. The results presented predicted the possible roles of some DE miRNAs in the induction of antiviral responses. The DE candidate miRNAs, including miR-146a, miR-146b, miR-205a, miR-205b and miR-449, can be investigated further for functional validation studies and to be used as novel prophylactic and therapeutic targets in tailoring or enhancing antiviral responses against AIV.
2020-08-04
Vaccines (publié)
doi.org
''COGITO in Space'': a thought experiment in exo-neurobiology
Daniela de Paulis
Stephen Whitmarsh
Robert Oostenveld
Guillaume Dumas
Michael Sanders
2020-08-03
(publié)
doi.org
SeroTracker: a global SARS-CoV-2 seroprevalence dashboard
Rahul K. Arora
Abel Joseph
Jordan Van Wyk
Simona Rocco
Austin Atmaja
Ewan May
Tingting Yan
Niklas Bobrovitz
Jonathan Chevrier
Matthew P. Cheng
Tyler Williamson
David L Buckeridge
2020-08-03
Lancet. Infectious Diseases (Print) (publié)
doi.org
Optimal Local and Remote Controllers With Unreliable Uplink Channels: An Elementary Proof
Mohammad Afshari
Aditya Mahajan
Recently, a model of a decentralized control system with local and remote controllers connected over unreliable channels was presented in [… (voir plus)1]. The model has a nonclassical information structure that is not partially nested. Nonetheless, it is shown in [1] that the optimal control strategies are linear functions of the state estimate (which is a nonlinear function of the observations). Their proof is based on a fairly sophisticated dynamic programming argument. In this article, we present an alternative and elementary proof of the result which uses common information-based conditional independence and completion of squares.
2020-07-31
IEEE Transactions on Automatic Control (publié)
doi.org
arxiv.org
Renewal Monte Carlo: Renewal Theory-Based Reinforcement Learning
Jayakumar Subramanian
Aditya Mahajan
An online reinforcement learning algorithm called renewal Monte Carlo (RMC) is presented. RMC works for infinite horizon Markov decision pro… (voir plus)cesses with a designated start state. RMC is a Monte Carlo algorithm that retains the key advantages of Monte Carlo—viz., simplicity, ease of implementation, and low bias—while circumventing the main drawbacks of Monte Carlo—viz., high variance and delayed updates. Given a parameterized policy
2020-07-31
IEEE Transactions on Automatic Control (publié)
doi.org
Towards Queryable and Traceable Domain Models
Rijul Saini
Gunter Mussbacher
Jin L.C. Guo
Jörg Kienzle
Model-Driven Software Engineering encompasses various modelling formalisms for supporting software development. One such formalism is domain… (voir plus) modelling which bridges the gap between requirements expressed in natural language and analyzable and more concise domain models expressed in class diagrams. Due to the lack of modelling skills among novice modellers and time constraints in industrial projects, it is often not possible to build an accurate domain model manually. To address this challenge, we aim to develop an approach to extract domain models from problem descriptions written in natural language by combining rules based on natural language processing with machine learning. As a first step, we report on an automated and tool-supported approach with an accuracy of extracted domain models higher than existing approaches. In addition, the approach generates trace links for each model element of a domain model. The trace links enable novice modellers to execute queries on the extracted domain models to gain insights into the modelling decisions taken for improving their modelling skills. Furthermore, to evaluate our approach, we propose a novel comparison metric and discuss our experimental design. Finally, we present a research agenda detailing research directions and discuss corresponding challenges.
2020-07-31
IEEE International Requirements Engineering Conference (publié)
doi.org
Inferring disease subtypes from clusters in explanation space
Marc-Andre Schulz
Matt Chapman-Rounds
Manisha Verma
Danilo Bzdok
Konstantinos Georgatzis
Identification of disease subtypes and corresponding biomarkers can substantially improve clinical diagnosis and treatment selection. Discov… (voir plus)ering these subtypes in noisy, high dimensional biomedical data is often impossible for humans and challenging for machines. We introduce a new approach to facilitate the discovery of disease subtypes: Instead of analyzing the original data, we train a diagnostic classifier (healthy vs. diseased) and extract instance-wise explanations for the classifier’s decisions. The distribution of instances in the explanation space of our diagnostic classifier amplifies the different reasons for belonging to the same class–resulting in a representation that is uniquely useful for discovering latent subtypes. We compare our ability to recover subtypes via cluster analysis on model explanations to classical cluster analysis on the original data. In multiple datasets with known ground-truth subclasses, particularly on UK Biobank brain imaging data and transcriptome data from the Cancer Genome Atlas, we show that cluster analysis on model explanations substantially outperforms the classical approach. While we believe clustering in explanation space to be particularly valuable for inferring disease subtypes, the method is more general and applicable to any kind of sub-type identification.
2020-07-29
Scientific Reports (publié)
doi.org
Patterns of autism symptoms: hidden structure in the ADOS and ADI-R instruments
Jeremy Lefort-Besnard
Kai Vogeley
Leonhard Schilbach
Gael Varoquaux
Bertrand Thirion
Guillaume Dumas
Danilo Bzdok
2020-07-29
Translational Psychiatry (publié)
doi.org
Deriving Differential Target Propagation from Iterating Approximate Inverses
Yoshua Bengio
2020-07-28
ArXiv (prépublication)
arxiv.org
arxiv.org
Predicting COVID-19 Pneumonia Severity on Chest X-ray With Deep Learning
Joseph Paul Cohen
Lan Dao
Paul Morrison
Karsten Roth
Yoshua Bengio
Beiyi Shen
Almas F Abbasi
Hoshmand Kochi Mahsa
Marzyeh Ghassemi
Haifang Li
Tim Q Duong
Introduction The need to streamline patient management for coronavirus disease-19 (COVID-19) has become more pressing than ever. Chest X-ray… (voir plus)s (CXRs) provide a non-invasive (potentially bedside) tool to monitor the progression of the disease. In this study, we present a severity score prediction model for COVID-19 pneumonia for frontal chest X-ray images. Such a tool can gauge the severity of COVID-19 lung infections (and pneumonia in general) that can be used for escalation or de-escalation of care as well as monitoring treatment efficacy, especially in the ICU. Methods Images from a public COVID-19 database were scored retrospectively by three blinded experts in terms of the extent of lung involvement as well as the degree of opacity. A neural network model that was pre-trained on large (non-COVID-19) chest X-ray datasets is used to construct features for COVID-19 images which are predictive for our task. Results This study finds that training a regression model on a subset of the outputs from this pre-trained chest X-ray model predicts our geographic extent score (range 0-8) with 1.14 mean absolute error (MAE) and our lung opacity score (range 0-6) with 0.78 MAE. Conclusions These results indicate that our model’s ability to gauge the severity of COVID-19 lung infections could be used for escalation or de-escalation of care as well as monitoring treatment efficacy, especially in the ICU. To enable follow up work, we make our code, labels, and data available online.
2020-07-27
Cureus (publié)
doi.org
arxiv.org