Because reinforcement learning suffers from a lack of scalability, online value (and Q-) function approximation has received increasing interest this last decade. This contribution introduces a novel approximation scheme, namely the Kalman Temporal Differences (KTD) framework, that exhibits the following features: sample-efficiency, non-linear approximation, non-stationarity handling and uncertainty management. A first KTD-based algorithm is provided for deterministic Markov Decision Processes (MDP) which produces biased estimates in the case of stochastic transitions. Than the eXtended KTD framework (XKTD), solving stochastic MDP, is described. Convergence is analyzed for special cases for both deterministic and stochastic transitions. Related algorithms are experimented on classical benchmarks. They compare favorably to the state of the art while exhibiting the announced features. 1.

Abstract. Reinforcement learning induces non-stationarity at several levels. Adaptation to non-stationary environments is of course a desired feature of a fair RL algorithm. Yet, even if the environment of the learning agent can be considered as stationary, generalized policy iteration frameworks, because of the interleaving of learning and control, will produce non-stationarity of the evaluated policy and so of its value function. Tracking the optimal solution instead of trying to converge to it is therefore preferable. In this paper, we propose to handle this tracking issue with a Kalman-based temporal difference framework. Complexity and convergence analysis are studied. Empirical investigations of its ability to handle non-stationarity is finally provided.

In several agent-oriented scenarios in the real world, an autonomous agent that is situated in an unknown environment must learn through a process of trial and error to take actions that result in long-term benefit. Reinforcement Learning (or sequential decision making) is a paradigm well-suited to this requirement. Value function-based methods and policy search methods are contrasting approaches to solve reinforcement learning tasks. While both classes of methods benefit from independent theoretical analyses, these often fail to extend to the practical situations in which the methods are deployed. We conduct an empirical study to examine the strengths and weaknesses of these approaches by introducing a suite of test domains that can be varied for problem size, stochasticity, function approximation, and partial observability. Our results indicate clear patterns in the domain characteristics for which each class of methods excels. We investigate whether their strengths can be combined, and develop an approach to achieve that purpose. The effectiveness of this approach is also demonstrated on the challenging benchmark task of robot soccer Keepaway. We highlight several lines of inquiry that emanate from this study.

Abstract. In reinforcement learning, the aim of the agent is to find a policy that maximizes its expected return. Policy gradient methods try to accomplish this goal by directly approximating the policy using a parametric function approximator; the expected return of the current policy is estimated and its parameters are updated by steepest ascent in the direction of the gradient of the expected return with respect to the policy parameters. In general, the policy is defined in terms of a set of basis functions that capture important features of the problem. Since the quality of the resulting policies directly depend on the set of basis functions, and defining them gets harder as the complexity of the problem increases, it is important to be able to find them automatically. In this paper, we propose a new approach which uses cascade-correlation learning architecture for automatically constructing a set of basis functions within the context of Natural Actor-Critic (NAC) algorithms. Such basis functions allow more complex policies be represented, and consequently improve the performance of the resulting policies. We also present the effectiveness of the method empirically. 1

Reinforcement learning for partially observable Markov decision problems (POMDPs) is a challenge as it requires policies with an internal state. Traditional approaches suffer significantly from this shortcoming and usually make strong assumptions on the problem domain such as perfect system models, state-estimators and a Markovian hidden system. Recurrent neural networks (RNNs) offer a natural framework for dealing with policy learning using hidden state and require only few limiting assumptions. As they can be trained well using gradient descent, they are suited for policy gradient approaches. In this paper, we present a policy gradient method, the Recurrent Policy Gradient which constitutes a model-free reinforcement learning method. It is aimed at training limited-memory stochastic policies on problems which require long-term memories of past observations. The approach involves approximating a policy gradient for a recurrent neural network by backpropagating return-weighted characteristic eligibilities through time. Using a “Long Short-Term Memory” RNN architecture, we are able to outperform previous RL methods on three important benchmark tasks. Furthermore, we show that using history-dependent baselines helps reducing estimation variance significantly, thus enabling our approach to tackle more challenging, highly stochastic environments.

Abstract not found

Actor-critic algorithms for reinforcement learning are achieving renewed popularity due to their good convergence properties in situations where other approaches often fail (e.g., when function approximation is involved). Interestingly, there is growing evidence that actor-critic approaches based on phasic dopamine signals play a key role in biological learning through cortical and basal ganglia loops. We derive a temporal difference based actor critic learning algorithm, for which convergence can be proved without assuming widely separated time scales for the actor and the critic. The approach is demonstrated by applying it to networks of spiking neurons. The established relation between phasic dopamine and the temporal difference signal lends support to the biological relevance of such algorithms. 1

This thesis examines reinforcement learning for stochastic control processes with single and multiple agents, where either the learning outcomes are stochastic policies or learning is perpetual and within the domain of stochastic policies. In this context, a policy is a strategy for processing environmental outputs (called observations) and subsequently generating a response or input-signal to the environment (called actions). A stochastic policy gives a probability distribution over actions for each observed situation, and the thesis concentrates on finite sets of observations and actions. There is an exclusive focus on stochastic policies for two principle reasons: such policies have been relatively neglected in the existing literature, and they have been recognised to be especially important in the field of multi-agent reinforcement learning. For the latter reason, the thesis concerns itself primarily with solutions best suited to multi-agent domains. This restriction proves essential, since the topic is otherwise too broad to be covered in depth without losing some clarity and focus. The thesis is partitioned into 3 parts, with chapter of contextual information preceding the first part. Part 1, focuses on analytic and formal mathematical approaches

Abstract — This paper presents an extension of existing cooperative control algorithms that have been developed for multi-UAV applications to utilize real-time observations and/or performance metric(s) in conjunction with learning methods to generate a more intelligent planner response. We approach this issue from a cooperative control perspective and embed elements of feedback control and active learning, resulting in an new intelligent Cooperative Control Architecture (iCCA). We describe this architecture, discuss some of the issues that must be addressed, and present illustrative examples of cooperative control problems where iCCA can be applied effectively. I.

Abstract—Reinforcement learning (RL) is a machine learning answer to the optimal control problem. It consists of learning an optimal control policy through interactions with the system to be controlled, the quality of this policy being quantified by the so-called value function. An important RL subtopic is to approximate this function when the system is too large for an exact representation. This survey reviews and unifies state of the art methods for parametric value function approximation by grouping them into three main categories: bootstrapping, residuals and projected fixed-point approaches. Related algorithms are derived by considering one of the associated cost functions and a specific way to minimize it, almost always a stochastic gradient descent or a recursive least-squares approach. Index Terms—Reinforcement learning, value function approximation, survey. I.