Portrait of Doina Precup

Doina Precup

Core Academic Member
Canada CIFAR AI Chair
Associate Professor, McGill University
Research Team Leader, Google DeepMind
Research Topics
Medical Machine Learning
Molecular Modeling
Probabilistic Models
Reasoning
Reinforcement Learning

Biography

Doina Precup combines teaching at McGill University with fundamental research on reinforcement learning, in particular AI applications in areas of significant social impact, such as health care. She is interested in machine decision-making in situations where uncertainty is high.

In addition to heading the Montreal office of Google DeepMind, Precup is a Senior Fellow of the Canadian Institute for Advanced Research and a Fellow of the Association for the Advancement of Artificial Intelligence.

Her areas of speciality are artificial intelligence, machine learning, reinforcement learning, reasoning and planning under uncertainty, and applications.

Current Students

Collaborating Alumni - McGill University
Co-supervisor :
Collaborating Alumni - McGill University
PhD - McGill University
Co-supervisor :
PhD - McGill University
Principal supervisor :
Master's Research - McGill University
Principal supervisor :
Master's Research - McGill University
Co-supervisor :
Master's Research - McGill University
Master's Research - McGill University
PhD - McGill University
PhD - Polytechnique Montréal
Principal supervisor :
Independent visiting researcher - McGill University
Co-supervisor :
Collaborating Alumni - McGill University
PhD - McGill University
Principal supervisor :
Master's Research - McGill University
Collaborating Alumni - McGill University
Principal supervisor :
PhD - McGill University
Principal supervisor :
Research Intern - McGill University
Master's Research - McGill University
Co-supervisor :
Research Intern - McGill University
PhD - McGill University
Principal supervisor :
PhD - McGill University
Collaborating Alumni - McGill University
Co-supervisor :

Publications

The Paradox of Choice: Using Attention in Hierarchical Reinforcement Learning
Appendix: On the Expressivity of Markov Reward
David Abel
Will Dabney
Anna Harutyunyan
Mark K. Ho
Michael L. Littman
Satinder Singh
(Q1) What does it mean for Bob to *solve* one of these tasks? That is, if Alice chooses a SOAP, PO, or TO for Bob to learn to solve, when ca… (see more)n Alice determine Bob has solved the task? A: Bob can be said to be doing better on a given task if his behavior improves, as is typical in evaluating behavior under reward. The difference with SOAPs, POs, and TOs is that we measure improvement relative to the task rather than reward. For instance, given a SOAP, we might say that Bob has solved the task once he has found one of the good policies, and we might measure Bob’s progress on a task in terms of the distance of his greedy policy to one of the good policies (as done in our learning experiments). The same reasoning applies to POs and TOs: Bob is doing better on a task in so far as his greedy policy (or trajectories) is (are) higher up the ordering.
Behind the Machine's Gaze: Biologically Constrained Neural Networks Exhibit Human-like Visual Attention
B. Eskofier
Dario Zanca
.
Low-Rank Representation of Reinforcement Learning Policies
We propose a general framework for policy representation for reinforcement learning tasks. This framework involves finding a low-dimensional… (see more) embedding of the policy on a reproducing kernel Hilbert space (RKHS). The usage of RKHS based methods allows us to derive strong theoretical guarantees on the expected return of the reconstructed policy. Such guarantees are typically lacking in black-box models, but are very desirable in tasks requiring stability and convergence guarantees. We conduct several experiments on classic RL domains. The results confirm that the policies can be robustly represented in a low-dimensional space while the embedded policy incurs almost no decrease in returns.
Importance of Empirical Sample Complexity Analysis for Offline Reinforcement Learning
Single-Shot Pruning for Offline Reinforcement Learning
Riyasat Ohib
Sergey Plis
A Consciousness-Inspired Planning Agent for Model-Based Reinforcement Learning
We present an end-to-end, model-based deep reinforcement learning agent which dynamically attends to relevant parts of its state during plan… (see more)ning. The agent uses a bottleneck mechanism over a set-based representation to force the number of entities to which the agent attends at each planning step to be small. In experiments, we investigate the bottleneck mechanism with several sets of customized environments featuring different challenges. We consistently observe that the design allows the planning agents to generalize their learned task-solving abilities in compatible unseen environments by attending to the relevant objects, leading to better out-of-distribution generalization performance.
Flexible Option Learning
Temporal abstraction in reinforcement learning (RL), offers the promise of improving generalization and knowledge transfer in complex enviro… (see more)nments, by propagating information more efficiently over time. Although option learning was initially formulated in a way that allows updating many options simultaneously, using off-policy, intra-option learning (Sutton, Precup & Singh, 1999), many of the recent hierarchical reinforcement learning approaches only update a single option at a time: the option currently executing. We revisit and extend intra-option learning in the context of deep reinforcement learning, in order to enable updating all options consistent with current primitive action choices, without introducing any additional estimates. Our method can therefore be naturally adopted in most hierarchical RL frameworks. When we combine our approach with the option-critic algorithm for option discovery, we obtain significant improvements in performance and data-efficiency across a wide variety of domains.
Flow Network based Generative Models for Non-Iterative Diverse Candidate Generation
This paper is about the problem of learning a stochastic policy for generating an object (like a molecular graph) from a sequence of actions… (see more), such that the probability of generating an object is proportional to a given positive reward for that object. Whereas standard return maximization tends to converge to a single return-maximizing sequence, there are cases where we would like to sample a diverse set of high-return solutions. These arise, for example, in black-box function optimization when few rounds are possible, each with large batches of queries, where the batches should be diverse, e.g., in the design of new molecules. One can also see this as a problem of approximately converting an energy function to a generative distribution. While MCMC methods can achieve that, they are expensive and generally only perform local exploration. Instead, training a generative policy amortizes the cost of search during training and yields to fast generation. Using insights from Temporal Difference learning, we propose GFlowNet, based on a view of the generative process as a flow network, making it possible to handle the tricky case where different trajectories can yield the same final state, e.g., there are many ways to sequentially add atoms to generate some molecular graph. We cast the set of trajectories as a flow and convert the flow consistency equations into a learning objective, akin to the casting of the Bellman equations into Temporal Difference methods. We prove that any global minimum of the proposed objectives yields a policy which samples from the desired distribution, and demonstrate the improved performance and diversity of GFlowNet on a simple domain where there are many modes to the reward function, and on a molecule synthesis task.
Gradient Starvation: A Learning Proclivity in Neural Networks
We identify and formalize a fundamental gradient descent phenomenon resulting in a learning proclivity in over-parameterized neural networks… (see more). Gradient Starvation arises when cross-entropy loss is minimized by capturing only a subset of features relevant for the task, despite the presence of other predictive features that fail to be discovered. This work provides a theoretical explanation for the emergence of such feature imbalance in neural networks. Using tools from Dynamical Systems theory, we identify simple properties of learning dynamics during gradient descent that lead to this imbalance, and prove that such a situation can be expected given certain statistical structure in training data. Based on our proposed formalism, we develop guarantees for a novel regularization method aimed at decoupling feature learning dynamics, improving accuracy and robustness in cases hindered by gradient starvation. We illustrate our findings with simple and real-world out-of-distribution (OOD) generalization experiments.
Temporally Abstract Partial Models
Humans and animals have the ability to reason and make predictions about different courses of action at many time scales. In reinforcement l… (see more)earning, option models (Sutton, Precup \& Singh, 1999; Precup, 2000) provide the framework for this kind of temporally abstract prediction and reasoning. Natural intelligent agents are also able to focus their attention on courses of action that are relevant or feasible in a given situation, sometimes termed affordable actions. In this paper, we define a notion of affordances for options, and develop temporally abstract partial option models, that take into account the fact that an option might be affordable only in certain situations. We analyze the trade-offs between estimation and approximation error in planning and learning when using such models, and identify some interesting special cases. Additionally, we demonstrate empirically the potential impact of partial option models on the efficiency of planning.
On the Expressivity of Markov Reward
David Abel
Will Dabney
Anna Harutyunyan
Mark K. Ho
Michael L. Littman
Satinder Singh
Reward is the driving force for reinforcement-learning agents. This paper is dedicated to understanding the expressivity of reward as a way … (see more)to capture tasks that we would want an agent to perform. We frame this study around three new abstract notions of"task"that might be desirable: (1) a set of acceptable behaviors, (2) a partial ordering over behaviors, or (3) a partial ordering over trajectories. Our main results prove that while reward can express many of these tasks, there exist instances of each task type that no Markov reward function can capture. We then provide a set of polynomial-time algorithms that construct a Markov reward function that allows an agent to optimize tasks of each of these three types, and correctly determine when no such reward function exists. We conclude with an empirical study that corroborates and illustrates our theoretical findings.