Score-based diffusion meets annealed importance sampling
More than twenty years after its introduction, Annealed Importance Sampling (AIS) remains
one of the most effective methods for marginal likelihood estimation. It relies on a sequence …
one of the most effective methods for marginal likelihood estimation. It relies on a sequence …
Mutual information regularized offline reinforcement learning
The major challenge of offline RL is the distribution shift that appears when out-of-
distribution actions are queried, which makes the policy improvement direction biased by …
distribution actions are queried, which makes the policy improvement direction biased by …
Probabilistic inference in language models via twisted sequential monte carlo
Numerous capability and safety techniques of Large Language Models (LLMs), including
RLHF, automated red-teaming, prompt engineering, and infilling, can be cast as sampling …
RLHF, automated red-teaming, prompt engineering, and infilling, can be cast as sampling …
Interpretable diffusion via information decomposition
Denoising diffusion models enable conditional generation and density modeling of complex
relationships like images and text. However, the nature of the learned relationships is …
relationships like images and text. However, the nature of the learned relationships is …
Provable benefits of annealing for estimating normalizing constants: Importance Sampling, Noise-Contrastive Estimation, and beyond
O Chehab, A Hyvarinen… - Advances in Neural …, 2024 - proceedings.neurips.cc
Recent research has developed several Monte Carlo methods for estimating the
normalization constant (partition function) based on the idea of annealing. This means …
normalization constant (partition function) based on the idea of annealing. This means …
Tight mutual information estimation with contrastive fenchel-legendre optimization
Abstract Successful applications of InfoNCE (Information Noise-Contrastive Estimation) and
its variants have popularized the use of contrastive variational mutual information (MI) …
its variants have popularized the use of contrastive variational mutual information (MI) …
Variational representations of annealing paths: Bregman information under monotonic embedding
R Brekelmans, F Nielsen - Information Geometry, 2024 - Springer
Abstract Markov chain Monte Carlo methods for sampling from complex distributions and
estimating normalization constants often simulate samples from a sequence of intermediate …
estimating normalization constants often simulate samples from a sequence of intermediate …
Attack-free evaluating and enhancing adversarial robustness on categorical data
Research on adversarial robustness has predominantly focused on continuous inputs,
leaving categorical inputs, especially tabular attributes, less examined. To echo this …
leaving categorical inputs, especially tabular attributes, less examined. To echo this …
Rho-tau bregman information and the geometry of annealing paths
R Brekelmans, F Nielsen - arXiv preprint arXiv:2209.07481, 2022 - arxiv.org
Markov Chain Monte Carlo methods for sampling from complex distributions and estimating
normalization constants often simulate samples from a sequence of intermediate …
normalization constants often simulate samples from a sequence of intermediate …
Optimizing the Noise in Self-Supervised Learning: from Importance Sampling to Noise-Contrastive Estimation
Self-supervised learning is an increasingly popular approach to unsupervised learning,
achieving state-of-the-art results. A prevalent approach consists in contrasting data points …
achieving state-of-the-art results. A prevalent approach consists in contrasting data points …