Reinforcement learning algorithms are gaining popularity in fields where optimal scheduling is important, and oncology is not an exception. The complex and uncertain dynamics of cancer limit the performance of traditional model-based scheduling strategies like Optimal Control. Motivated by the recent success of model-free Deep Reinforcement Learning (DRL) in challenging control tasks and in medical treatments, we use Deep Q-Network (DQN) and Deep Deterministic Policy Gradient (DDPG) to design a personalized cancer chemotherapy schedule. We show that both of them succeed in the task and outperform the Optimal Control solution in the presence of uncertainty. Furthermore, we show that DDPG can exterminate cancer more efficiently than DQN due to its continuous action space. Finally, we provide some intuition regarding the amount of samples required for the training.

Imitation learning is an effective alternative approach to learn a policy when the reward function is sparse. In this paper, we consider a challenging setting where an agent and an expert use different actions from each other. We assume that the agent has access to a sparse reward function and state-only expert observations. We propose a method which gradually balances between the imitation learning cost and the reinforcement learning objective. In addition, this method adapts the agent's policy based on either mimicking expert behavior or maximizing sparse reward. We show, through navigation scenarios, that (i) an agent is able to efficiently leverage sparse rewards to outperform standard state-only imitation learning, (ii) it can learn a policy even when its actions are different from the expert, and (iii) the performance of the agent is not bounded by that of the expert, due to the optimized usage of sparse rewards.

Attention models have had a significant positive impact on deep learning across a range of tasks. However previous attempts at integrating attention with reinforcement learning have failed to produce significant improvements. We propose the first combination of self attention and reinforcement learning that is capable of producing significant improvements, including new state of the art results in the Arcade Learning Environment. Unlike the selective attention models used in previous attempts, which constrain the attention via preconceived notions of importance, our implementation utilises the Markovian properties inherent in the state input. Our method produces a faithful visualisation of the policy, focusing on the behaviour of the agent. Our experiments demonstrate that the trained policies use multiple simultaneous foci of attention, and are able to modulate attention over time to deal with situations of partial observability.

Researchers on artificial intelligence have achieved human-level intelligence in large-scale perfect-information games, but it is still a challenge to achieve (nearly) optimal results (in other words, an approximate Nash Equilibrium) in large-scale imperfect-information games (i.e. war games, football coach or business strategies). Neural Fictitious Self Play (NFSP) is an effective algorithm for learning approximate Nash equilibrium of imperfect-information games from self-play without prior domain knowledge. However, it relies on Deep Q-Network, which is off-line and is hard to converge in online games with changing opponent strategy, so it can't approach approximate Nash equilibrium in games with large search scale and deep search depth. In this paper, we propose Monte Carlo Neural Fictitious Self Play (MC-NFSP), an algorithm combines Monte Carlo tree search with NFSP, which greatly improves the performance on large-scale zero-sum imperfect-information games. Experimentally, we demonstrate that the proposed Monte Carlo Neural Fictitious Self Play can converge to approximate Nash equilibrium in games with large-scale search depth while the Neural Fictitious Self Play can't. Furthermore, we develop Asynchronous Neural Fictitious Self Play (ANFSP). It use asynchronous and parallel architecture to collect game experience. In experiments, we show that parallel actor-learners have a further accelerated and stabilizing effect on training.

Deep reinforcement learning (deep RL) has achieved superior performance in complex sequential tasks by using a deep neural network as its function approximator and by learning directly from raw images. A drawback of using raw images is that deep RL must learn the state feature representation from the raw images in addition to learning a policy. As a result, deep RL can require a prohibitively large amount of training time and data to reach reasonable performance, making it difficult to use deep RL in real-world applications, especially when data is expensive. In this work, we speed up training by addressing half of what deep RL is trying to solve --- learning features. Our approach is to learn some of the important features by pre-training deep RL network's hidden layers via supervised learning using a small set of human demonstrations. We empirically evaluate our approach using deep Q-network (DQN) and asynchronous advantage actor-critic (A3C) algorithms on the Atari 2600 games of Pong, Freeway, and Beamrider. Our results show that: 1) pre-training with human demonstrations in a supervised learning manner is better at discovering features relative to pre-training naively in DQN, and 2) initializing a deep RL network with a pre-trained model provides a significant improvement in training time even when pre-training from a small number of human demonstrations.

Policy gradient algorithms typically combine discounted future rewards with an estimated value function, to compute the direction and magnitude of parameter updates. However, for most Reinforcement Learning tasks, humans can provide additional insight to constrain the policy learning. We introduce a general method to incorporate multiple different feedback channels into a single policy gradient loss. In our formulation, the Multi-Preference Actor Critic (M-PAC), these different types of feedback are implemented as constraints on the policy. We use a Lagrangian relaxation to satisfy these constraints using gradient descent while learning a policy that maximizes rewards. Experiments in Atari and Pendulum verify that constraints are being respected and can accelerate the learning process.

The ability to transfer in reinforcement learning is key towards building an agent of general artificial intelligence. In this paper, we consider the problem of learning to simultaneously transfer across both environments (ENV) and tasks (TASK), probably more importantly, by learning from only sparse (ENV, TASK) pairs out of all the possible combinations. We propose a novel compositional neural network architecture which depicts a meta rule for composing policies from the environment and task embeddings. Notably, one of the main challenges is to learn the embeddings jointly with the meta rule. We further propose new training methods to disentangle the embeddings, making them both distinctive signatures of the environments and tasks and effective building blocks for composing the policies. Experiments on GridWorld and Thor, of which the agent takes as input an egocentric view, show that our approach gives rise to high success rates on all the (ENV, TASK) pairs after learning from only 40\% of them.

A key stepping stone in the development of an artificial general intelligence (a machine that can perform any task), is the production of agents that can perform multiple tasks at once instead of just one. Unfortunately, canonical methods are very prone to catastrophic forgetting (CF) - the act of overwriting previous knowledge about a task when learning a new task. Recent efforts have developed techniques for overcoming CF in learning systems, but no attempt has been made to apply these new techniques to evolutionary systems. This research presents a novel technique, weight protection, for reducing CF in evolutionary systems by adapting a method from learning systems. It is used in conjunction with other evolutionary approaches for overcoming CF and is shown to be effective at alleviating CF when applied to a suite of reinforcement learning tasks. It is speculated that this work could indicate the potential for a wider application of existing learning-based approaches to evolutionary systems and that evolutionary techniques may be competitive with or better than learning systems when it comes to reducing CF.

Physical construction -- the ability to compose objects, subject to physical dynamics, in order to serve some function -- is fundamental to human intelligence. Here we introduce a suite of challenging physical construction tasks inspired by how children play with blocks, such as matching a target configuration, stacking and attaching blocks to connect objects together, and creating shelter-like structures over target objects. We then examine how a range of modern deep reinforcement learning agents fare on these challenges, and introduce several new approaches which provide superior performance. Our results show that agents which use structured representations (e.g., objects and scene graphs) and structured policies (e.g., object-centric actions) outperform those which use less structured representations, and generalize better beyond their training when asked to reason about larger scenes. Agents which use model-based planning via Monte-Carlo Tree Search also outperform strictly model-free agents in our most challenging construction problems. We conclude that approaches which combine structured representations and reasoning with powerful learning are a key path toward agents that possess rich intuitive physics, scene understanding, and planning.

A long-standing challenge in Reinforcement Learning is enabling agents to learn a model of their environment which can be transferred to solve other problems in a world with the same underlying rules. One reason this is difficult is the challenge of learning accurate models of an environment. If such a model is inaccurate, the agent's plans and actions will likely be sub-optimal, and likely lead to the wrong outcomes. Recent progress in model-based reinforcement learning has improved the ability for agents to learn and use predictive models. In this paper, we extend a recent deep learning architecture which learns a predictive model of the environment that aims to predict only the value of a few key measurements, which are be indicative of an agent's performance. Predicting only a few measurements rather than the entire future state of an environment makes it more feasible to learn a valuable predictive model. We extend this predictive model with a small, evolving neural network that suggests the best goals to pursue in the current state. We demonstrate that this allows the predictive model to transfer to new scenarios where goals are different, and that the adaptive goals can even adjust agent behavior on-line, changing its strategy to fit the current context.