A game-theoretic perspective of deep neural networks

Article Management

Selfishness Need Not Be Bad

Zijun Wu et al.

Summary: This study investigates the price of anarchy in nonatomic congestion games when the total demand gets very large, extending previous results in this direction and offering a new framework for the limit analysis of PoA. The research shows that regardless of the growth pattern of total demand T, PoA converges to one in the limit game.

OPERATIONS RESEARCH (2021)

Add to Collection

Article Computer Science, Artificial Intelligence

Explainable deep learning for efficient and robust pattern recognition: A survey of recent developments

Xiao Bai et al.

Summary: This article introduces the success of deep learning in visual recognition tasks, but mentions the black-box problem, the difficulty of understanding decisions, and restrictions on safety-critical applications. It also outlines the three main categories of explainable deep learning methods, efficient deep learning via model compression and acceleration, and robustness and stability in deep learning, presenting representative works and latest developments.

PATTERN RECOGNITION (2021)

Add to Collection

Article Computer Science, Artificial Intelligence