NLP in the era of generative AI, cognitive sciences, and societal transformation
Join us at Mila in October for a three-day workshop to explore the transformative potential of language technologies and their implications for society.
This program is designed to provide decision-makers, policymakers and professional working in policy with a foundational understanding of AI technology.
We use cookies to analyze the browsing and usage of our website and to personalize your experience. You can disable these technologies at any time, but this may limit certain functionalities of the site. Read our Privacy Policy for more information.
Setting cookies
You can enable and disable the types of cookies you wish to accept. However certain choices you make could affect the services offered on our sites (e.g. suggestions, personalised ads, etc.).
Essential cookies
These cookies are necessary for the operation of the site and cannot be deactivated. (Still active)
Analytics cookies
Do you accept the use of cookies to measure the audience of our sites?
Multimedia Player
Do you accept the use of cookies to display and allow you to watch the video content hosted by our partners (YouTube, etc.)?
Tail averaging consists in averaging the last examples in a stream. Common techniques either have a memory requirement which grows with the … (see more)number of samples to average, are not available at every timestep or do not accomodate growing windows. We propose two techniques with a low constant memory cost that perform tail averaging with access to the average at every time step. We also show how one can improve the accuracy of that average at the cost of increased memory consumption.
Despite many algorithmic advances, our theoretical understanding of practical distributional reinforcement learning methods remains limited.… (see more) One exception is Rowland et al. (2018)'s analysis of the C51 algorithm in terms of the Cramer distance, but their results only apply to the tabular setting and ignore C51's use of a softmax to produce normalized distributions. In this paper we adapt the Cramer distance to deal with arbitrary vectors. From it we derive a new distributional algorithm which is fully Cramer-based and can be combined to linear function approximation, with formal guarantees in the context of policy evaluation.
In allowing the model's prediction to be any real vector, we lose the probabilistic interpretation behind the method, but otherwise maintain the appealing properties of distributional approaches. To the best of our knowledge, ours is the first proof of convergence of a distributional algorithm combined with function approximation. Perhaps surprisingly, our results provide evidence that Cramer-based distributional methods may perform worse than directly approximating the value function.
Despite many algorithmic advances, our theoretical understanding of practical distributional reinforcement learning methods remains limited.… (see more) One exception is Rowland et al. (2018)'s analysis of the C51 algorithm in terms of the Cramer distance, but their results only apply to the tabular setting and ignore C51's use of a softmax to produce normalized distributions. In this paper we adapt the Cramer distance to deal with arbitrary vectors. From it we derive a new distributional algorithm which is fully Cramer-based and can be combined to linear function approximation, with formal guarantees in the context of policy evaluation.
In allowing the model's prediction to be any real vector, we lose the probabilistic interpretation behind the method, but otherwise maintain the appealing properties of distributional approaches. To the best of our knowledge, ours is the first proof of convergence of a distributional algorithm combined with function approximation. Perhaps surprisingly, our results provide evidence that Cramer-based distributional methods may perform worse than directly approximating the value function.
We propose a new perspective on representation learning in reinforcement learning based on geometric properties of the space of value functi… (see more)ons. We leverage this perspective to provide formal evidence regarding the usefulness of value functions as auxiliary tasks. Our formulation considers adapting the representation to minimize the (linear) approximation of the value function of all stationary policies for a given environment. We show that this optimization reduces to making accurate predictions regarding a special class of value functions which we call adversarial value functions (AVFs). We demonstrate that using value functions as auxiliary tasks corresponds to an expected-error relaxation of our formulation, with AVFs a natural candidate, and identify a close relationship with proto-value functions (Mahadevan, 2005). We highlight characteristics of AVFs and their usefulness as auxiliary tasks in a series of experiments on the four-room domain.
Online communities such as Facebook and Twitter are enormously popular and have become an essential part of the daily life of many of their … (see more)users. Through these platforms, users can discover and create information that others will then consume. In that context, recommending relevant information to users becomes critical for viability. However, recommendation in online communities is a challenging problem: 1) users' interests are dynamic, and 2) users are influenced by their friends. Moreover, the influencers may be context-dependent. That is, different friends may be relied upon for different topics. Modeling both signals is therefore essential for recommendations. We propose a recommender system for online communities based on a dynamic-graph-attention neural network. We model dynamic user behaviors with a recurrent neural network, and context-dependent social influence with a graph-attention neural network, which dynamically infers the influencers based on users' current interests. The whole model can be efficiently fit on large-scale data. Experimental results on several real-world data sets demonstrate the effectiveness of our proposed approach over several competitive baselines including state-of-the-art models.
2019-01-30
Proceedings of the Twelfth ACM International Conference on Web Search and Data Mining (published)
Existing approaches to automatic summarization assume that a length limit for the summary is given, and view content selection as an optimiz… (see more)ation problem to maximize informativeness and minimize redundancy within this budget. This framework ignores the fact that human-written summaries have rich internal structure which can be exploited to train a summarization system. We present NEXTSUM, a novel approach to summarization based on a model that predicts the next sentence to include in the summary using not only the source article, but also the summary produced so far. We show that such a model successfully captures summary-specific discourse moves, and leads to better content selection performance, in addition to automatically predicting how long the target summary should be. We perform experiments on the New York Times Annotated Corpus of summaries, where NEXTSUM outperforms lead and content-model summarization baselines by significant margins. We also show that the lengths of summaries produced by our system correlates with the lengths of the human-written gold standards.
We propose a new perspective on representation learning in reinforcement learning based on geometric properties of the space of value functi… (see more)ons. We leverage this perspective to provide formal evidence regarding the usefulness of value functions as auxiliary tasks. Our formulation considers adapting the representation to minimize the (linear) approximation of the value function of all stationary policies for a given environment. We show that this optimization reduces to making accurate predictions regarding a special class of value functions which we call adversarial value functions (AVFs). We demonstrate that using value functions as auxiliary tasks corresponds to an expected-error relaxation of our formulation, with AVFs a natural candidate, and identify a close relationship with proto-value functions (Mahadevan, 2005). We highlight characteristics of AVFs and their usefulness as auxiliary tasks in a series of experiments on the four-room domain.
Policy-makers’ efforts to increase the uptake of advance medical directives (AMDs), and the legal constraints they impose on health profes… (see more)sionals, are bringing greater scrutiny to provincial AMD regimes. In 2015, Quebec introduced a new, legally binding form to be filled out for AMDs, which limits individuals’ expression of their wishes to narrow, checklist responses to questions on specific medical interventions. This form-focused regime has other shortcomings: it relies on individuals to self-inform and it does not provide them the opportunity to meaningfully convey their preferences for end-of-life care. A more values-based and collaborative approach provides a better path forward for Quebec and for other provinces.
Adversarial robustness has become a central goal in deep learning, both in theory and practice. However, successful methods to improve adver… (see more)sarial robustness (such as adversarial training) greatly hurt generalization performance on the clean data. This could have a major impact on how adversarial robustness affects real world systems (i.e. many may opt to forego robustness if it can improve performance on the clean data). We propose Interpolated Adversarial Training, which employs recently proposed interpolation based training methods in the framework of adversarial training. On CIFAR-10, adversarial training increases clean test error from 5.8% to 16.7%, whereas with our Interpolated adversarial training we retain adversarial robustness while achieving a clean test error of only 6.5%. With our technique, the relative error increase for the robust model is reduced from 187.9% to just 12.1%.
Adversarial robustness has become a central goal in deep learning, both in theory and practice. However, successful methods to improve adver… (see more)sarial robustness (such as adversarial training) greatly hurt generalization performance on the clean data. This could have a major impact on how adversarial robustness affects real world systems (i.e. many may opt to forego robustness if it can improve performance on the clean data). We propose Interpolated Adversarial Training, which employs recently proposed interpolation based training methods in the framework of adversarial training. On CIFAR-10, adversarial training increases clean test error from 5.8% to 16.7%, whereas with our Interpolated adversarial training we retain adversarial robustness while achieving a clean test error of only 6.5%. With our technique, the relative error increase for the robust model is reduced from 187.9% to just 12.1%.