Jun Sakuma

NeurIPS Conference 2023 Conference Paper

Demographic Parity Constrained Minimax Optimal Regression under Linear Model

• Kazuto Fukuchi
• Jun Sakuma

We explore the minimax optimal error associated with a demographic parity-constrained regression problem within the context of a linear model. Our proposed model encompasses a broader range of discriminatory bias sources compared to the model presented by Chzhen and Schreuder. Our analysis reveals that the minimax optimal error for the demographic parity-constrained regression problem under our model is characterized by $\Theta(\frac{dM}{n})$, where $n$ denotes the sample size, $d$ represents the dimensionality, and $M$ signifies the number of demographic groups arising from sensitive attributes. Moreover, we demonstrate that the minimax error increases in conjunction with a larger bias present in the model.

IJCAI Conference 2023 Conference Paper

Statistically Significant Concept-based Explanation of Image Classifiers via Model Knockoffs

• Kaiwen Xu
• Kazuto Fukuchi
• Youhei Akimoto
• Jun Sakuma

A concept-based classifier can explain the decision process of a deep learning model by human understandable concepts in image classification problems. However, sometimes concept-based explanations may cause false positives, which misregards unrelated concepts as important for the prediction task. Our goal is to find the statistically significant concept for classification to prevent misinterpretation. In this study, we propose a method using a deep learning model to learn the image concept and then using the knockoff sample to select the important concepts for prediction by controlling the False Discovery Rate (FDR) under a certain value. We evaluate the proposed method in our experiments on both synthetic and real data. Also, it shows that our method can control the FDR properly while selecting highly interpretable concepts to improve the trustworthiness of the model.

NeurIPS Conference 2022 Conference Paper

Max-Min Off-Policy Actor-Critic Method Focusing on Worst-Case Robustness to Model Misspecification

• Takumi Tanabe
• Rei Sato
• Kazuto Fukuchi
• Jun Sakuma
• Youhei Akimoto

In the field of reinforcement learning, because of the high cost and risk of policy training in the real world, policies are trained in a simulation environment and transferred to the corresponding real-world environment. However, the simulation environment does not perfectly mimic the real-world environment, lead to model misspecification. Multiple studies report significant deterioration of policy performance in a real-world environment. In this study, we focus on scenarios involving a simulation environment with uncertainty parameters and the set of their possible values, called the uncertainty parameter set. The aim is to optimize the worst-case performance on the uncertainty parameter set to guarantee the performance in the corresponding real-world environment. To obtain a policy for the optimization, we propose an off-policy actor-critic approach called the Max-Min Twin Delayed Deep Deterministic Policy Gradient algorithm (M2TD3), which solves a max-min optimization problem using a simultaneous gradient ascent descent approach. Experiments in multi-joint dynamics with contact (MuJoCo) environments show that the proposed method exhibited a worst-case performance superior to several baseline approaches.

AAAI Conference 2022 Conference Paper

Unsupervised Causal Binary Concepts Discovery with VAE for Black-Box Model Explanation

• Thien Q Tran
• Kazuto Fukuchi
• Youhei Akimoto
• Jun Sakuma

We aim to explain a black-box classifier with the form: ‘data X is classified as class Y because X has A, B and does not have C’ in which A, B, and C are high-level concepts. The challenge is that we have to discover in an unsupervised manner a set of concepts, i. e. , A, B and C, that is useful for explaining the classifier. We first introduce a structural generative model that is suitable to express and discover such concepts. We then propose a learning process that simultaneously learns the data distribution and encourages certain concepts to have a large causal influence on the classifier output. Our method also allows easy integration of user’s prior knowledge to induce high interpretability of concepts. Finally, using multiple datasets, we demonstrate that the proposed method can discover useful concepts for explanation in this form.

AAAI Conference 2021 Conference Paper

AdvantageNAS: Efficient Neural Architecture Search with Credit Assignment

• Rei Sato
• Jun Sakuma
• Youhei Akimoto

Neural architecture search (NAS) is an approach for automatically designing a neural network architecture without human effort or expert knowledge. However, the high computational cost of NAS limits its use in commercial applications. Two recent NAS paradigms, namely one-shot and sparse propagation, which reduce the time and space complexities, respectively, provide clues for solving this problem. In this paper, we propose a novel search strategy for one-shot and sparse propagation NAS, namely AdvantageNAS, which further reduces the time complexity of NAS by reducing the number of search iterations. AdvantageNAS is a gradientbased approach that improves the search efficiency by introducing credit assignment in gradient estimation for architecture updates. Experiments on the NAS-Bench-201 and PTB dataset show that AdvantageNAS discovers an architecture with higher performance under a limited time budget compared to existing sparse propagation NAS. To further reveal the reliabilities of AdvantageNAS, we investigate it theoretically and find that it monotonically improves the expected loss and thus converges.

AAAI Conference 2020 Conference Paper

Generate (Non-Software) Bugs to Fool Classifiers

• Hiromu Yakura
• Youhei Akimoto
• Jun Sakuma

In adversarial attacks intended to confound deep learning models, most studies have focused on limiting the magnitude of the modification so that humans do not notice the attack. On the other hand, during an attack against autonomous cars, for example, most drivers would not find it strange if a small insect image were placed on a stop sign, or they may overlook it. In this paper, we present a systematic approach to generate natural adversarial examples against classification models by employing such natural-appearing perturbations that imitate a certain object or signal. We first show the feasibility of this approach in an attack against an image classifier by employing generative adversarial networks that produce image patches that have the appearance of a natural object to fool the target model. We also introduce an algorithm to optimize placement of the perturbation in accordance with the input image, which makes the generation of adversarial examples fast and likely to succeed. Moreover, we experimentally show that the proposed approach can be extended to the audio domain, for example, to generate perturbations that sound like the chirping of birds to fool a speech classifier.

IJCAI Conference 2019 Conference Paper

Robust Audio Adversarial Example for a Physical Attack

• Hiromu Yakura
• Jun Sakuma

We propose a method to generate audio adversarial examples that can attack a state-of-the-art speech recognition model in the physical world. Previous work assumes that generated adversarial examples are directly fed to the recognition model, and is not able to perform such a physical attack because of reverberation and noise from playback environments. In contrast, our method obtains robust adversarial examples by simulating transformations caused by playback or recording in the physical world and incorporating the transformations into the generation process. Evaluation and a listening experiment demonstrated that our adversarial examples are able to attack without being noticed by humans. This result suggests that audio adversarial examples generated by the proposed method may become a real threat.

AAAI Conference 2018 Conference Paper

Efficiently Monitoring Small Data Modification Effect for Large-Scale Learning in Changing Environment

• Hiroyuki Hanada
• Atsushi Shibagaki
• Jun Sakuma
• Ichiro Takeuchi

We study large-scale machine learning problems in changing environments where a small part of the dataset is modified, and the effect of the data modification must be monitored in order to know how much the modification changes the optimal model. When the entire dataset is large, even if the amount of the data modification is fairly small, the computational cost for re-training the model would be prohibitively large. In this paper, we propose a novel method, called the optimal solution bounding (OSB), for monitoring such a data modification effect on the optimal model by efficiently evaluating (without actually re-training) it. The proposed method provides bounds on the unknown optimal model with the cost proportional only to the size of the data modification.

ICML Conference 2017 Conference Paper

Differentially Private Chi-squared Test by Unit Circle Mechanism

• Kazuya Kakizaki
• Kazuto Fukuchi
• Jun Sakuma

This paper develops differentially private mechanisms for $\chi^2$ test of independence. While existing works put their effort into properly controlling the type-I error, in addition to that, we investigate the type-II error of differentially private mechanisms. Based on the analysis, we present unit circle mechanism: a novel differentially private mechanism based on the geometrical property of the test statistics. Compared to existing output perturbation mechanisms, our mechanism improves the dominated term of the type-II error from $O(1)$ to $O(\exp(-\sqrt{N}))$ where $N$ is the sample size. Furthermore, we introduce novel procedures for multiple $\chi^2$ tests by incorporating the unit circle mechanism into the sparse vector technique and the exponential mechanism. These procedures can control the family-wise error rate (FWER) properly, which has never been attained by existing mechanisms.

ICML Conference 2010 Conference Paper

Online Prediction with Privacy

• Jun Sakuma
• Hiromi Arai

ICML Conference 2008 Conference Paper

Privacy-preserving reinforcement learning

• Jun Sakuma
• Shigenobu Kobayashi
• Rebecca N. Wright

We consider the problem of distributed reinforcement learning (DRL) from private perceptions. In our setting, agents' perceptions, such as states, rewards, and actions, are not only distributed but also should be kept private. Conventional DRL algorithms can handle multiple agents, but do not necessarily guarantee privacy preservation and may not guarantee optimality. In this work, we design cryptographic solutions that achieve optimal policies without requiring the agents to share their private information.