Joelle Pineau

(Rex) Devon Hjelm

Exploration is a crucial component for discovering approximately optimal policies in most high-dimensional reinforcement learning (RL) setti… (see more)ngs with sparse rewards. Approaches such as neural density models and continuous exploration (e.g., Go-Explore) have been instrumental in recent advances. Soft actor-critic (SAC) is a method for improving exploration that aims to combine off-policy updates while maximizing the policy entropy. We extend SAC to a richer class of probability distributions through normalizing flows, which we show improves performance in exploration, sample complexity, and convergence. Finally, we show that not only the normalizing flow policy outperforms SAC on MuJoCo domains, it is also significantly lighter, using as low as 5.6% of the original network's parameters for similar performance.

2020-05-12

Proceedings of the Conference on Robot Learning (published)

Literature Mining for Incorporating Inductive Bias in Biomedical Prediction Tasks (Student Abstract)

Qizhen Zhang

2020-04-03

AAAI Conference on Artificial Intelligence (published)

doi.org

Provably efficient reconstruction of policy networks

Bogdan Mazoure

Thang Doan

Tianyu Li

Recent research has shown that learning poli-cies parametrized by large neural networks can achieve significant success on challenging reinf… (see more)orcement learning problems. However, when memory is limited, it is not always possible to store such models exactly for inference, and com-pressing the policy into a compact representation might be necessary. We propose a general framework for policy representation, which reduces this problem to finding a low-dimensional embedding of a given density function in a separable inner product space. Our framework allows us to de-rive strong theoretical guarantees, controlling the error of the reconstructed policies. Such guaran-tees are typically lacking in black-box models, but are very desirable in risk-sensitive tasks. Our experimental results suggest that the reconstructed policies can use less than 10%of the number of parameters in the original networks, while incurring almost no decrease in rewards.

2020-02-07

ArXiv (preprint)

Representation of Reinforcement Learning Policies in Reproducing Kernel Hilbert Spaces.

Bogdan Mazoure

Thang Doan

Tianyu Li

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. We conduct several experiments on classic RL domains. The results confirm that the policies can be robustly embedded in a low-dimensional space while the embedded policy incurs almost no decrease in return.

2020-02-07

(published)

www.semanticscholar.org

Language GANs Falling Short

Massimo Caccia

Lucas Caccia

William Fedus

Hugo Larochelle

Laurent Charlin

Generating high-quality text with sufficient diversity is essential for a wide range of Natural Language Generation (NLG) tasks. Maximum-Lik… (see more)elihood (MLE) models trained with teacher forcing have consistently been reported as weak baselines, where poor performance is attributed to exposure bias (Bengio et al., 2015; Ranzato et al., 2015); at inference time, the model is fed its own prediction instead of a ground-truth token, which can lead to accumulating errors and poor samples. This line of reasoning has led to an outbreak of adversarial based approaches for NLG, on the account that GANs do not suffer from exposure bias. In this work, we make several surprising observations which contradict common beliefs. First, we revisit the canonical evaluation framework for NLG, and point out fundamental flaws with quality-only evaluation: we show that one can outperform such metrics using a simple, well-known temperature parameter to artificially reduce the entropy of the model's conditional distributions. Second, we leverage the control over the quality / diversity trade-off given by this parameter to evaluate models over the whole quality-diversity spectrum and find MLE models constantly outperform the proposed GAN variants over the whole quality-diversity space. Our results have several implications: 1) The impact of exposure bias on sample quality is less severe than previously thought, 2) temperature tuning provides a better quality / diversity trade-off than adversarial training while being easier to train, easier to cross-validate, and less computationally expensive. Code to reproduce the experiments is available at github.com/pclucas14/GansFallingShort

2020-01-01

ICLR (published)

openreview.net

Attraction-Repulsion Actor-Critic for Continuous Control Reinforcement Learning

Thang Doan

Bogdan Mazoure

(Rex) Devon Hjelm

Continuous control tasks in reinforcement learning are important because they provide an important framework for learning in high-dimensiona… (see more)l state spaces with deceptive rewards, where the agent can easily become trapped into suboptimal solutions. One way to avoid local optima is to use a population of agents to ensure coverage of the policy space, yet learning a population with the "best" coverage is still an open problem. In this work, we present a novel approach to population-based RL in continuous control that leverages properties of normalizing flows to perform attractive and repulsive operations between current members of the population and previously observed policies. Empirical results on the MuJoCo suite demonstrate a high performance gain for our algorithm compared to prior work, including Soft-Actor Critic (SAC).

2019-09-17

ArXiv (preprint)

openreview.net

Online Adaptative Curriculum Learning for GANs

Thang Doan

Joao Monteiro

Isabela Albuquerque

Bogdan Mazoure

(Rex) Devon Hjelm

Generative Adversarial Networks (GANs) can successfully approximate a probability distribution and produce realistic samples. However, open … (see more)questions such as sufficient convergence conditions and mode collapse still persist. In this paper, we build on existing work in the area by proposing a novel framework for training the generator against an ensemble of discriminator networks, which can be seen as a one-student/multiple-teachers setting. We formalize this problem within the full-information adversarial bandit framework, where we evaluate the capability of an algorithm to select mixtures of discriminators for providing the generator with feedback during learning. To this end, we propose a reward function which reflects the progress made by the generator and dynamically update the mixture weights allocated to each discriminator. We also draw connections between our algorithm and stochastic optimization methods and then show that existing approaches using multiple discriminators in literature can be recovered from our framework. We argue that less expressive discriminators are smoother and have a general coarse grained view of the modes map, which enforces the generator to cover a wide portion of the data distribution support. On the other hand, highly expressive discriminators ensure samples quality. Finally, experimental results show that our approach improves samples quality and diversity over existing baselines by effectively learning a curriculum. These results also support the claim that weaker discriminators have higher entropy improving modes coverage.

2019-07-17

Proceedings of the AAAI Conference on Artificial Intelligence (published)

doi.org

Multitask Metric Learning: Theory and Algorithm

Boyu Wang

Hejia Zhang

Peng Liu

Zebang Shen

In this paper, we study the problem of multitask metric learning (mtML). We first examine the generalization bound of the regularized mtML f… (see more)ormulation based on the notion of algorithmic stability, proving the convergence rate of mtML and revealing the trade-off between the tasks. Moreover, we also establish the theoretical connection between the mtML, single-task learning and pooling-task learning approaches. In addition, we present a novel boosting-based mtML (mt-BML) algorithm, which scales well with the feature dimension of the data. Finally, we also devise an efficient second-order Riemannian retraction operator which is tailored specifically to our mt-BML algorithm. It produces a low-rank solution of mtML to reduce the model complexity, and may also improve generalization performances. Extensive evaluations on several benchmark data sets verify the effectiveness of our learning algorithm.

2019-04-11

International Conference on Artificial Intelligence and Statistics (published)

dblp.uni-trier.de

Multitask Metric Learning: Theory and Algorithm

Boyu Wang

Hejia Zhang

Peng Liu

Zebang Shen

2019-04-11

International Conference on Artificial Intelligence and Statistics (published)

Contextual Bandits for Adapting Treatment in a Mouse Model of de Novo Carcinogenesis

Charis Achilleos

Demetris C Iacovides

Katerina Strati

Georgios D. Mitsis

In this work, we present a specific case study where we aim to design effective treatment allocation strategies and validate these using a m… (see more)ouse model of skin cancer. Collecting data for modelling treatments effectiveness on animal models is an expensive and time consuming process. Moreover, acquiring this information during the full range of disease stages is hard to achieve with a conventional random treatment allocation procedure, as poor treatments cause deterioration of subject health. We therefore aim to design an adaptive allocation strategy to improve the efficiency of data collection by allocating more samples for exploring promising treatments. We cast this application as a contextual bandit problem and introduce a simple and practical algorithm for exploration-exploitation in this framework. The work builds on a recent class of approaches for non-contextual bandits that relies on subsampling to compare treatment options using an equivalent amount of information. On the technical side, we extend the subsampling strategy to the case of bandits with context, by applying subsampling within Gaussian Process regression. On the experimental side, preliminary results using 10 mice with skin tumours suggest that the proposed approach extends by more than 50% the subjects life duration compared with baseline strategies: no treatment, random treatment allocation, and constant chemotherapeutic agent. By slowing the tumour growth rate, the adaptive procedure gathers information about treatment effectiveness on a broader range of tumour volumes, which is crucial for eventually deriving sequential pharmacological treatment strategies for cancer.

2018-11-29

Proceedings of the 3rd Machine Learning for Healthcare Conference (published)

Focused Hierarchical RNNs for Conditional Sequence Processing

Nan Rosemary Ke

Konrad Żołna

Alessandro Sordoni

Zhouhan Lin

Adam Trischler

Recurrent Neural Networks (RNNs) with attention mechanisms have obtained state-of-the-art results for many sequence processing tasks. Most o… (see more)f these models use a simple form of encoder with attention that looks over the entire sequence and assigns a weight to each token independently. We present a mechanism for focusing RNN encoders for sequence modelling tasks which allows them to attend to key parts of the input as needed. We formulate this using a multi-layer conditional sequence encoder that reads in one token at a time and makes a discrete decision on whether the token is relevant to the context or question being asked. The discrete gating mechanism takes in the context embedding and the current hidden state as inputs and controls information flow into the layer above. We train it using policy gradient methods. We evaluate this method on several types of tasks with different attributes. First, we evaluate the method on synthetic tasks which allow us to evaluate the model for its generalization ability and probe the behavior of the gates in more controlled settings. We then evaluate this approach on large scale Question Answering tasks including the challenging MS MARCO and SearchQA tasks. Our models shows consistent improvements for both tasks over prior work and our baselines. It has also shown to generalize significantly better on synthetic tasks as compared to the baselines.

2018-07-03

Proceedings of the 35th International Conference on Machine Learning (published)

RE-EVALUATE: Reproducibility in Evaluating Reinforcement Learning Algorithms

Khimya Khetarpal

Zafarali Ahmed

Andre Cianflone

Riashat Islam