This work introduces HACK GPs (Hedge Adaptive Cumulative Kernels), a method that views kernel selection as an online learning with expert advice problem and establishes general guarantees showing that, under a loss-gap condition, the weight concentrates on the best kernel and the resulting acquisition function is close...
Kavin Aravindan, Mani Tej Sriram, Gautam Dasarathy et al.· Performance evaluation (Prin...· 0 citations
This work proposes an algorithm, TOFU-POV, that estimates the latent action subspace using the masked actions, imputes current actions using an epoch-wise frozen representation, and runs OFUL in the resulting low-dimensional coordinates, and devise a rank-adaptive algorithm that does not require the knowledge of the in...
Gautam Dasarathy, V. Gattani, Lalit Jain· arXiv.org· 0 citations
This analysis tackles the problem of exact recovery of the graph from one trajectory of random-scan Gaussian Glauber dynamics by demonstrating how to extract fresh Gaussian innovations from the update sequence, which yields mixing-free control of appropriate quantities.
