Takuya Kanazawa

Takuya Kanazawa, Haiyan Wang, Chetan Gupta

Uncertainty quantification is one of the central challenges for machine learning in real-world applications. In reinforcement learning, an agent confronts two kinds of uncertainty, called epistemic uncertainty and aleatoric uncertainty. Disentangling and evaluating these uncertainties simultaneously stands a chance of improving the agent's final performance, accelerating training, and facilitating quality assurance after deployment. In this work, we propose an uncertainty-aware reinforcement learning algorithm for continuous control tasks that extends the Deep Deterministic Policy Gradient algorithm (DDPG). It exploits epistemic uncertainty to accelerate exploration and aleatoric uncertainty to learn a risk-sensitive policy. We conduct numerical experiments showing that our variant of DDPG outperforms vanilla DDPG without uncertainty estimation in benchmark tasks on robotic control and power-grid optimization.

Takuya Kanazawa, Chetan Gupta

Development of an accurate, flexible, and numerically efficient uncertainty quantification (UQ) method is one of fundamental challenges in machine learning. Previously, a UQ method called DISCO Nets has been proposed (Bouchacourt et al., 2016), which trains a neural network by minimizing the energy score. In this method, a random noise vector in $\mathbb{R}^{10\text{--}100}$ is concatenated with the original input vector in order to produce a diverse ensemble forecast despite using a single neural network. While this method has shown promising performance on a hand pose estimation task in computer vision, it remained unexplored whether this method works as nicely for regression on tabular data, and how it competes with more recent advanced UQ methods such as NGBoost. In this paper, we propose an improved neural architecture of DISCO Nets that admits faster and more stable training while only using a compact noise vector of dimension $\sim \mathcal{O}(1)$. We benchmark this approach on miscellaneous real-world tabular datasets and confirm that it is competitive with or even superior to standard UQ baselines. Moreover we observe that it exhibits better point forecast performance than a neural network of the same size trained with the conventional mean squared error. As another advantage of the proposed method, we show that local feature importance computation methods such as SHAP can be easily applied to any subregion of the predictive distribution. A new elementary proof for the validity of using the energy score to learn predictive distributions is also provided.

Takuya Kanazawa, Chetan Gupta

Sequential decision making in the real world often requires finding a good balance of conflicting objectives. In general, there exist a plethora of Pareto-optimal policies that embody different patterns of compromises between objectives, and it is technically challenging to obtain them exhaustively using deep neural networks. In this work, we propose a novel multi-objective reinforcement learning (MORL) algorithm that trains a single neural network via policy gradient to approximately obtain the entire Pareto set in a single run of training, without relying on linear scalarization of objectives. The proposed method works in both continuous and discrete action spaces with no design change of the policy network. Numerical experiments in benchmark environments demonstrate the practicality and efficacy of our approach in comparison to standard MORL baselines.

Takuya Kanazawa

Quantifying predictive uncertainty is essential for safe and trustworthy real-world AI deployment. Yet, fully nonparametric estimation of conditional distributions remains challenging for multivariate targets. We propose Tomographic Quantile Forests (TQF), a nonparametric, uncertainty-aware, tree-based regression model for multivariate targets. TQF learns conditional quantiles of directional projections $\mathbf{n}^{\top}\mathbf{y}$ as functions of the input $\mathbf{x}$ and the unit direction $\mathbf{n}$. At inference, it aggregates quantiles across many directions and reconstructs the multivariate conditional distribution by minimizing the sliced Wasserstein distance via an efficient alternating scheme with convex subproblems. Unlike classical directional-quantile approaches that typically produce only convex quantile regions and require training separate models for different directions, TQF covers all directions with a single model without imposing convexity restrictions. We evaluate TQF on synthetic and real-world datasets, and release the source code on GitHub.

Takuya Kanazawa

Bayesian optimization (BO) with Gaussian processes is a powerful methodology to optimize an expensive black-box function with as few function evaluations as possible. The expected improvement (EI) and probability of improvement (PI) are among the most widely used schemes for BO. There is a plethora of other schemes that outperform EI and PI, but most of them are numerically far more expensive than EI and PI. In this work, we propose a new one-parameter family of acquisition functions for BO that unifies EI and PI. The proposed method is numerically inexpensive, is easy to implement, can be easily parallelized, and on benchmark tasks shows a performance superior to EI and GP-UCB. Its generalization to BO with Student-t processes is also presented.

Takuya Kanazawa

Bayesian optimization is a powerful tool to optimize a black-box function, the evaluation of which is time-consuming or costly. In this paper, we propose a new approach to Bayesian optimization called GP-MGC, which maximizes multiscale graph correlation with respect to the global maximum to determine the next query point. We present our evaluation of GP-MGC in applications involving both synthetic benchmark functions and real-world datasets and demonstrate that GP-MGC performs as well as or even better than state-of-the-art methods such as max-value entropy search and GP-UCB.

Takuya Kanazawa

The need to collect data via expensive measurements of black-box functions is prevalent across science, engineering and medicine. As an example, hyperparameter tuning of a large AI model is critical to its predictive performance but is generally time-consuming and unwieldy. Bayesian optimization (BO) is a collection of methods that aim to address this issue by means of Bayesian statistical inference. In this work, we put forward a BO scheme named BDC, which integrates BO with a statistical measure of association of two random variables called Distance Correlation. BDC balances exploration and exploitation automatically, and requires no manual hyperparameter tuning. We evaluate BDC on a range of benchmark tests and observe that it performs on per with popular BO methods such as the expected improvement and max-value entropy search. We also apply BDC to optimization of sequential integral observations of an unknown terrain and confirm its utility.

Takuya Kanazawa, Akinori Asahara, Hidekazu Morita

Making material experiments more efficient is a high priority for materials scientists who seek to discover new materials with desirable properties. In this paper, we investigate how to optimize the laborious sequential measurements of materials properties with data-driven methods, taking the small-angle neutron scattering (SANS) experiment as a test case. We propose two methods for optimizing sequential data sampling. These methods iteratively suggest the best target for the next measurement by performing a statistical analysis of the already acquired data, so that maximal information is gained at each step of an experiment. We conducted numerical simulations of SANS experiments for virtual materials and confirmed that the proposed methods significantly outperform baselines.