Hirotaka Hachiya

Importance-weighted least-squares probabilistic classifier for covariate shift adaptation with application to human activity recognition

Human activity recognition from accelerometer data (e.g., obtained by smart phones) is gathering a great deal of attention since it can be used for various purposes such as remote health-care. However, since collecting labeled data is bothersome for new users, it is desirable to utilize data obtained from existing users. In this paper, we formulate this adaptation problem as learning under covariate shift, and propose a computationally efficient probabilistic classification method based on adaptive importance sampling. The usefulness of the proposed method is demonstrated in real-world human activity recognition.

Adaptive importance sampling for value function approximation in off-policy reinforcement learning

Off-policy reinforcement learning is aimed at efficiently using data samples gathered from a policy that is different from the currently optimized policy. A common approach is to use importance sampling techniques for compensating for the bias of value function estimators caused by the difference between the data-sampling policy and the target policy. However, existing off-policy methods often do not take the variance of the value function estimators explicitly into account and therefore their performance tends to be unstable. To cope with this problem, we propose using an adaptive importance sampling technique which allows us to actively control the trade-off between bias and variance. We further provide a method for optimally determining the trade-off parameter based on a variant of cross-validation. We demonstrate the usefulness of the proposed approach through simulations.

Analysis and improvement of policy gradient estimation

Policy gradient is a useful model-free reinforcement learning approach, but it tends to suffer from instability of gradient estimates. In this paper, we analyze and improve the stability of policy gradient methods. We first prove that the variance of gradient estimates in the PGPE (policy gradients with parameter-based exploration) method is smaller than that of the classical REINFORCE method under a mild assumption. We then derive the optimal baseline for PGPE, which contributes to further reducing the variance. We also theoretically show that PGPE with the optimal baseline is more preferable than REINFORCE with the optimal baseline in terms of the variance of gradient estimates. Finally, we demonstrate the usefulness of the improved PGPE method through experiments.

Geodesic Gaussian kernels for value function approximation

The least-squares policy iteration approach works efficiently in value function approximation, given appropriate basis functions. Because of its smoothness, the Gaussian kernel is a popular and useful choice as a basis function. However, it does not allow for discontinuity which typically arises in real-world reinforcement learning tasks. In this paper, we propose a new basis function based on geodesic Gaussian kernels, which exploits the non-linear manifold structure induced by the Markov decision processes. The usefulness of the proposed method is successfully demonstrated in simulated robot arm control and Khepera robot navigation.

Feature selection for reinforcement learning: evaluating implicit state-reward dependency via conditional mutual information

Model-free reinforcement learning (RL) is a machine learning approach to decision making in unknown environments. However, realworld RL tasks often involve high-dimensional state spaces, and then standard RL methods do not perform well. In this paper, we propose a new feature selection framework for coping with high dimensionality. Our proposed framework adopts conditional mutual information between return and state-feature sequences as a feature selection criterion, allowing the evaluation of implicit state-reward dependency. The conditional mutual information is approximated by a least-squares method, which results in a computationally efficient feature selection procedure. The usefulness of the proposed method is demonstrated on grid-world navigation problems.

Efficient sample reuse in policy gradients with parameter-based exploration

The policy gradient approach is a flexible and powerful reinforcement learning method particularly for problems with continuous actions such as robot control. A common challenge is how to reduce the variance of policy gradient estimates for reliable policy updates. In this letter, we combine the following three ideas and give a highly effective policy gradient method: (1) policy gradients with parameter-based exploration, a recently proposed policy search method with low variance of gradient estimates; (2) an importance sampling technique, which allows us to reuse previously gathered data in a consistent way; and (3) an optimal baseline, which minimizes the variance of gradient estimates with their unbiasedness being maintained. For the proposed method, we give a theoretical analysis of the variance of gradient estimates and show its usefulness through extensive experiments.

Information-Maximization Clustering Based on Squared-Loss Mutual Information

Information-maximization clustering learns a probabilistic classifier in an unsupervised manner so that mutual information between feature vectors and cluster assignments is maximized. A notable advantage of this approach is that it involves only continuous optimization of model parameters, which is substantially simpler than discrete optimization of cluster assignments. However, existing methods still involve nonconvex optimization problems, and therefore finding a good local optimal solution is not straightforward in practice. In this letter, we propose an alternative information-maximization clustering method based on a squared-loss variant of mutual information. This novel approach gives a clustering solution analytically in a computationally efficient way via kernel eigenvalue decomposition. Furthermore, we provide a practical model selection procedure that allows us to objectively optimize tuning parameters included in the kernel function. Through experiments, we demonstrate the usefulness of the proposed approach.

Efficient exploration through active learning for value function approximation in reinforcement learning

Appropriately designing sampling policies is highly important for obtaining better control policies in reinforcement learning. In this paper, we first show that the least-squares policy iteration (LSPI) framework allows us to employ statistical active learning methods for linear regression. Then we propose a design method of good sampling policies for efficient exploration, which is particularly useful when the sampling cost of immediate rewards is high. The effectiveness of the proposed method, which we call active policy iteration (API), is demonstrated through simulations with a batting robot.

Reward-weighted regression with sample reuse for direct policy search in reinforcement learning

Direct policy search is a promising reinforcement learning framework, in particular for controlling continuous, high-dimensional systems. Policy search often requires a large number of samples for obtaining a stable policy update estimator, and this is prohibitive when the sampling cost is expensive. In this letter, we extend an expectation-maximization-based policy search method so that previously collected samples can be efficiently reused. The usefulness of the proposed method, reward-weighted regression with sample reuse (R), is demonstrated through robot learning experiments. (This letter is an extended version of our earlier conference paper: Hachiya, Peters, & Sugiyama, 2009.)

Artist Agent: A Reinforcement Learning Approach to Automatic Stroke Generation in Oriental Ink Painting

Oriental ink painting, called Sumi-e, is one of the most appealing painting styles that has attracted artists around the world. Major challenges in computer-based Sumi-e simulation are to abstract complex scene information and draw smooth and natural brush strokes. To automatically generate such strokes, we propose to model a brush as a reinforcement learning agent, and learn desired brush-trajectories by maximizing the sum of rewards in the policy search framework. We also elaborate on the design of actions, states, and rewards tailored for a Sumi-e agent. The effectiveness of our proposed approach is demonstrated through simulated Sumi-e experiments.