Many high-dimensional datasets concentrate near a low-dimensional structure embedded in the ambient space. Generative models for such data must control off-support mass while remaining computationally practical. Diffusion models use iterative denoising at inference, whereas standard normalizing flows require invertible...
A universal approximation theorem is established for DFNNs, advancing the state-of-the-art of neural network approximation theory to general metric-space-valued responses, without making model assumptions or relying on local smoothing.
Kyum Kim, Yaqing Chen, Paromita Dubey· arXiv.org· 3 citations
This monograph studies the relations between two approaches using positive definite kernels: probabilistic methods using Gaussian processes, and non-probabilistic methods using reproducing kernel Hilbert spaces (RKHS). They are widely studied and used in machine learning, statistics, and numerical analysis. We study co...
Motonobu Kanagawa, Philipp Hennig, Dino Sejdinovic et al.· 1 citation
An overview of the theory of overparameterized ML is provided by focusing on explaining the most foundational findings through a statistical signal processing perspective and emphasizing the unique aspects that define the TOPML research area as a subfield of modern ML theory and outline interesting open frontiers that...
Yehuda Dar, Vidya K. Muthukumar, Richard Baraniuk· arXiv.org· 83 citations· ⚡4
Reach audiences
Advertise in front of researchers, engineers, and readers.
This work proposes a regularized maximum likelihood framework for simultaneous parameter estimation and feature selection in MoE whose experts belong to the generalized linear model family, covering Gaussian, Poisson and multinomial responses within a single formulation.
Thin Nguyen-Van, Faicel Chamroukhi, Ha Hoang Van et al.· 15 citations
Heterogeneous cardiac signals can mutually supervise a single foundation model of cardiac physiology using a physiology-aware joint-embedding predictive architecture built on a physiology-aware joint-embedding predictive architecture.
Hamza Shafiq, Hung Manh Pham, Bin Zhu et al.· 3 citations· ⚡1
The central result recovers the KL order as the decay rate of the directional Fisher quadratic form approaching the singularity, in original coordinates, without a Hironaka resolution.
This paper develops new variance-reduction techniques for the forward-reflected-backward splitting (FRBS) method to solve a class of possibly nonmonotone stochastic composite inclusions. Unlike unbiased estimators such as mini-batching, developing stochastic biased variants faces a fundamental technical challenge and h...
This paper offers theoretical insights into when MCR yields a more favorable finite-sample estimation-error upper bound, viewed through the lens of neural network distance, and proposes a novel practical diagnostic that leverages the duality gap and a calibrated drift estimator to infer the potential benefit of MCR tra...
A simple decomposition of a neural-network-based normalizing flow that naturally uncovers a pivotal statistic in the presence of nuisance parameters, based only on a sample generator from the distribution of interest is presented.
Two new nuisances are derived as fixed points of a self-induced Bellman equation, in which they approximate the maximum Bellman operator by its softmax correspondence, and a debiased estimator is proposed through the Neyman orthogonality and established its asymptotic normality under diverging horizons.
Nan Lu, Ethan Lee, James M. Robins et al.· 0 citations
Although probabilistic statements are ubiquitous, foundational disagreements persist about their understanding, as exemplified by debates between Bayesians and frequentists; moreover, it is unclear when and why acting on them actually leads to desirable outcomes. Here, we argue that every probability is the output of a...
Learn how Google DeepMind and GovTech Singapore uses lightweight on-device Gemma models to intercept conversational scams in private chats with zero cloud data egress.
MIT News · Artificial Intelligence· news.mit.eduOct 8, 2026
