Masked diffusion models have shown promising performance in generating high-quality samples in a wide range of domains, but accelerating their sampling process remains relatively underexplored. To investigate efficient samplers for masked diffusion, this paper theoretically analyzes the MaskGIT sampler for image modeli...
Satoshi Hayakawa, Yuhta Takida, Masaaki Imaizumi et al.· 0 citations
This paper proposes a novel three-step aggregation method that first aligns the permutations across all local estimators via a particular spectral clustering approach, then aligns the signs within each estimated cluster via another tailored spectral approach, and finally applies the geometric median for robust aggregat...
The message is that the statistical cost of depth is the metric entropy of the set $B(k,F)$ of compositions, and it separates the models whose estimation error is independent of the depth from those whose estimation error grows with it.
Sho Sonoda, Yuka Hashimoto, Isao Ishikawa et al.· 0 citations
In multi-agent reinforcement learning (MARL) settings, agents are often heterogeneous with diverse intrinsic utility functions. In this study, we introduce Latent Objective Discovery (LOD) as a framework for decomposing agents' distinct utilities into mixed underlying objectives in a low-dimensional latent space. Laten...
Ke Sun, Zhaoyang Shi· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
We apply entropy to reinforcement learning in two ways: entropy-augmented reward defines the soft objective, while relative-entropy regularization controls the size of a policy-improvement step without changing that objective. Together they yield a one-parameter family of advanced policies connecting a base policy to i...
We study two-arm contextual bandits with arm-specific single indices and a shared unknown monotone link. Monotonicity makes the optimal action depend only on the contrast between the index directions, hence arm-specific reward functions need not be estimated. We introduce Natural Boundary Learning (NBL), a greedy proce...
We introduce the Functional Spectral-Newton Method (FSNM) for learning the leading singular structure of a conditional expectation operator without fixing a basis or reproducing kernel Hilbert space. FSNM fits a low-rank representation of the centered joint-to-product density ratio kernel by alternating functional Newt...
Thiago Ramos, A. Frohlich, Daniela Perazzo et al.· 0 citations
MTL-CMO is introduced, a multi-task framework that jointly learns shared function spaces and task-specific operators across multiple datasets and T-CMO, a transfer learning method that reuses the shared spaces to estimate, in closed form, the operator of a new conditional distribution.
Sami Chemlal, Thibaut Germain, Rémi Flamary et al.· 0 citations
This paper studies convex optimization when the gradient cannot be evaluated exactly, but only approximated by a hierarchy of algorithms whose compute grows like $\delta^{-\gamma}$ in the accuracy $\delta$. When $\gamma>2$, falling into the Harder-Than-Monte-Carlo (HTMC) regime, the price of accuracy outruns the varian...
A hierarchy of three entropy-based Shapley games makes this output-side choice explicit for any ordered multivariate outcome, ranging from per-component marginal entropy to fully joint entropy, and establishes a chain-rule decomposition of the joint attribution and characterize the cross-component term through conditio...
Niklas Koenen, Claudia Battistin, Jeriek Van den Abeele et al.· 0 citations
GUIDE-FBO is proposed, in which agents exchange compact distributions over the locations of their respective optima inferred from local Gaussian process (GP) posteriors, rather than raw observations, query points, or surrogate parameters, and remains effective across settings ranging from homogeneous to severely hetero...
Tabular foundation models (TFMs) provide predictive distributions for regression, but their prediction regions can exhibit undercoverage or overcoverage even when point predictions are accurate. We introduce C-USIM (Conditionally-Uniformized Score Integration Method), a lightweight application of highest predictive den...
Sungwoo Park, Sunghee Park, Won Chang· 0 citations
Jennifer Neville did not want to go into computer science—but that’s exactly where she landed. Neville discusses the starts and stops that led to her professional sweet spot and her work identifying “surprising failures” making it hard for AI to handle complexity. The post What AI gets wrong and what failure teaches us appeared first on Microsoft Research.
