Multivariate ordinal data along with covariates are commonly collected in problems ranging from alignment of language models with human preferences, as well as in recommender systems. For example, data sets such as MovieLens contain several movies rated on a scale 1--5 by human users, along with their demographic infor...
Yujie Chen, Antik Chakraborty, Anindya Bhadra· 0 citations
This work studies the problem of learning transition kernels for time-homogeneous jump-diffusion processes using conditional diffusion models and establishes non-asymptotic bounds for the conditional score estimation error and for the KL divergence between the laws of the true and generated discretely observed paths.
Weak-to-strong generalization is the phenomenon where a strong student model trained with labels produced by a weak teacher model is able to generalize better than the teacher. In this paper, we study this phenomenon in two-layer random feature networks where the model strength is determined by its width. Using tools f...
Finding low-energy configurations of interacting particles and approximating probability distributions lead to the minimization of potential-interaction energies in Wasserstein space. These energies can be nonconvex, making it important to exploit second-order information while controlling the reliability of local appr...
You Wan, Ting Gao, Jinqiao Duan· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
We study community recovery in the sparse symmetric stochastic block model with $q$ communities, average degree $d$ and signal strength $\lambda$ through statistical decision theory and Fisher information, and obtain three characterizations of the Kesten-Stigum threshold $d\lambda^2=1$ and of the information-computatio...
Machine unlearning asks for a deletion algorithm whose output is close to retraining from scratch without the selected forget examples. In this work, we study forget-only unlearning, where the deletion algorithm receives only the trained model and the examples to forget, with no retained data or extra training informat...
Luka Radi\'c, Vikrant Singhal, Amartya Sanyal· 0 citations
This work presents the first oracle-efficient algorithm that achieves sublinear regret in the agnostic setting without knowledge of $\mu$, based on Gaussian Follow-The-Perturbed-Leader, which is parameter-free and establishes the regret bound.
S. Voitovych, Adam Block, Alexander Rakhlin et al.· 2 citations
Two sampling methods are proposed, Size-Corrected 2LS (S-2LS) and Size- and Dispersion-Corrected 2LS (SD-2LS), which provide provably better softmax approximations with negligible to non-existent computational overhead.
This work introduces an observation-corrected Koopman framework for accelerating frozen diffusion models, and identifies finite-dimensional, time-dependent Koopman approximations that jointly describe the increments of shallow and deep network features.
Modern applications of AI rely on increasingly complex models. Explainable AI (XAI) has emerged as a set of techniques aimed at improving model transparency. However, existing XAI methods typically assume input features to be inherently interpretable, or they rely on intermediate internal abstractions that are difficul...
T. Schnake, Doreen Schöppenthau, Alexander Meyer et al.· 0 citations
What can a distribution-matching regularizer such as SIGReg in LeJEPA certify about contrastive learning? We study shared Gaussianization (SG), a characteristic-function Gaussianity test on the average of two normalized views, scaled by an independent $\chi_d$ radius. Because disagreeing views shorten the average, one...
Ruo-Yu Zhao, Yu-Ting Chen, Jin-He Zhang et al.· 0 citations
Wasserstein gradient flow extends gradient descent to probability measures. Its Hessian-guided perturbed variant (PWGF) adds Gaussian perturbations to escape saddle points in nonconvex problems. We investigate when its approximation by finitely many interacting particles remains accurate over growing time horizons. Our...
A machine-learning approach called CheMeleon provides a practical way to develop new drugs or other molecules with much less detailed, expensive experimentation.
Is your company’s AI sovereign? A practical guide for CIOs on AI vendor lock-in, EU Data Act switching, cloud sovereignty, and testing your exit plan before you need it. The post Sovereignty Is the Option to Leave appeared first on GPT-Lab.
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
