30
55
30
30
The Surprising Effectiveness of Approximate Value Iteration in Self-Play
Minimal approximate value iteration self-play learns more accurate value functions than AlphaZero in Connect Four and Hex while cutting training and inference costs.
The paper trains a minimal self-play implementation of Approximate Value Iteration (AVI) without MCTS and uses ground-truth oracles for exact evaluation in Connect Four, 7x7 Hex, and synthetic games. AVI learns more accurate value functions than AlphaZero, and its one-step-lookahead greedy policies remain competitive with MCTS-based policies at substantially lower training and inference cost. Preliminary experiments on Othello and 9x9 Go show AVI trains stably on larger games, suggesting simpler approaches have become increasingly practical with modern deep-learning tools.
35
30
30
57
60
42
30
42
60
60
30
30
Personal Data of Approximately 220,000 Domestic and International Gangnam Unni Users Leaked
Healing Paper's Gangnam Unni beauty platform leaked personal data of roughly 220,000 domestic and international users via abnormal API access on September 4.
Healing Paper, operator of the beauty medical platform Gangnam Unni, announced on September 7 that personal information of approximately 220,000 domestic and international users was leaked. The incident involved abnormal access on September 4 to the integration API used to view consultation records. The company disclosed the event through a public notice.
35
57
57
42
30
30
55
45
60
30
30
55
57
30
Shadow AI, deepfakes, and supply chain compromise are rewriting the financial sector threat playbook
60
57
45
57
60
30
55
45
60
42