Deciding When to Switch: E-Processes for Adaptive Minimax Training for Generative Adversarial Nets

The document is an arXiv RSS collection of newly announced statistical machine learning and causal inference research papers. Topics include adaptive sequential testing for GAN training, representation geometry, spectral embeddings, temporal-difference learning inference, conditional independence testing, ensemble early stopping, black-box prediction inference, random ellipsoid feasibility thresholds, causal adjustment scores, and neural representation distances. No cybersecurity vulnerability or exploit is described.