Discovering diverse solutions in deep reinforcement learning by maximizing state-action-based mutual information.

Reinforcement learning algorithms are typically limited to learning a single solution for a specified task, even though diverse solutions often exist. Recent studies showed that learning a set of diverse solutions is beneficial because diversity enables robust few-shot adaptation. Although existing methods learn diverse solutions by using the mutual information as unsupervised rewards, such an approach often suffers from the bias of the gradient estimator induced by value function approximation.

Sufficient Dimension Reduction via Direct Estimation of the Gradients of Logarithmic Conditional Densities.

Sufficient dimension reduction (SDR) is aimed at obtaining the low-rank projection matrix in the input space such that information about output data is maximally preserved. Among various approaches to SDR, a promising method is based on the eigendecomposition of the outer product of the gradient of the conditional density of output given input. In this letter, we propose a novel estimator of the gradient of the logarithmic conditional density that directly fits a linear-in-parameter model to the true gradient under the squared loss.

Direct Estimation of the Derivative of Quadratic Mutual Information with Application in Supervised Dimension Reduction.

A typical goal of linear-supervised dimension reduction is to find a low-dimensional subspace of the input space such that the projected input variables preserve maximal information about the output variables. The dependence-maximization approach solves the supervised dimension-reduction problem through maximizing a statistical dependence between projected input variables and output variables. A well-known statistical dependence measure is mutual information (MI), which is based on the Kullback-Leibler (KL) divergence.

Model-based reinforcement learning with dimension reduction.

The goal of reinforcement learning is to learn an optimal policy which controls an agent to acquire the maximum cumulative reward. The model-based reinforcement learning approach learns a transition model of the environment from data, and then derives the optimal policy using the transition model. However, learning an accurate transition model in high-dimensional environments requires a large amount of data which is difficult to obtain.

Conditional density estimation with dimensionality reduction via squared-loss conditional entropy minimization.

Regression aims at estimating the conditional mean of output given input. However, regression is not informative enough if the conditional density is multimodal, heteroskedastic, and asymmetric. In such a case, estimating the conditional density itself is preferable, but conditional density estimation (CDE) is challenging in high-dimensional space.

Model-based policy gradients with parameter-based exploration by least-squares conditional density estimation.

The goal of reinforcement learning (RL) is to let an agent learn an optimal control policy in an unknown environment so that future expected rewards are maximized. The model-free RL approach directly learns the policy based on data samples. Although using many samples tends to improve the accuracy of policy learning, collecting a large number of samples is often expensive in practice.

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.