Inference Algorithm

Inference algorithms aim to efficiently estimate the underlying parameters or hidden states of a system given observed data, a crucial task across diverse scientific fields. Current research emphasizes improving the speed and accuracy of inference, particularly for complex models like large language models and Bayesian networks, focusing on techniques such as distributed speculative inference, variational inference, and approximate message passing. These advancements are driving progress in areas ranging from material science (GISAXS data analysis) to human-computer interaction (activity recognition) by enabling more efficient and accurate analysis of increasingly complex datasets. Furthermore, research is actively exploring the interplay between data characteristics, model architecture, and the inference algorithm itself to optimize performance.

Papers

September 19, 2024

Unrolled denoising networks provably learn optimal Bayesian inference
Aayush Karan, Kulin Shah, Sitan Chen, Yonina C. Eldar
Bayesian Inference Compressed Sensing Inference Algorithm Informative Prior Supervised Counterpart

May 23, 2024

Distributed Speculative Inference of Large Language Models is Provably Faster
Nadav Timor, Jonathan Mamou, Daniel Korat, Moshe Berchansky, Oren Pereg, Moshe Wasserblat, Tomer Galanti, Michal Gordon, David Harel
Large Language Model Scientific Inference Autoregressive Model Inference Algorithm Speculative Exploration

February 10, 2024

Efficient Incremental Belief Updates Using Weighted Virtual Observations
David Tolpin
Inference Algorithm Probabilistic Program Iterative Inference Belief Update Virtual Measurement

December 15, 2023

Automatic Rao-Blackwellization for Sequential Monte Carlo with Belief Propagation
Waïss Azizian, Guillaume Baudart, Marc Lelarge
Belief Propagation Sequential Monte Carlo Inference Algorithm Rao Blackwellized Exact Bayesian Inference

October 27, 2023

Deep Transformed Gaussian Processes
Francisco Javier Sáez-Maldonado, Juan Maroñas, Daniel Hernández-Lobato
Variational Inference Deep Gaussian Process Inference Algorithm Transformed Gaussian Process

September 20, 2023

Colour Passing Revisited: Lifted Model Construction with Commutative Factors
Malte Luttermann, Tanya Braun, Ralf Möller, Marcel Gehrke
Scientific Inference Inference Algorithm Lifted Inference Tractable Probabilistic Inference Probabilistic Inference Exploit Symmetry

June 6, 2023

Learning to Simulate Tree-Branch Dynamics for Manipulation
Jayadeep Jacob, Tirthankar Bandyopadhyay, Jason Williams, Paulo Borges, Fabio Ramos
LeArning Abstract Strategic Manipulation Inference Algorithm Deformable Terrain 3D Tree Partial Inference

April 24, 2023

Human intuition as a defense against attribute inference
Marcin Waniek, Navya Suri, Abdullah Zameek, Bedoor AlShebli, Talal Rahwan
Artificial Intelligence Private Data Hidden Knowledge Human Intuition Inference Algorithm Attribute Inference

March 21, 2023

Efficient Multi-stage Inference on Tabular Data
Daniel S Johnson, Igor L Markov
End to End Tabular Data Machine Learning System Inference Latency Real Time Inference Inference Algorithm Step Inference

February 1, 2023

Automatically Marginalized MCMC in Probabilistic Programming
Jinlin Lai, Javier Burroni, Hui Guan, Daniel Sheldon
Markov Chain Monte Carlo Graphical Model Hamiltonian Monte Carlo Bayesian Model Probabilistic Programming Inference Algorithm Marginalization Model

November 11, 2022

Reconstruction of gene regulatory network via sparse optimization
Jiashu Lou, Leyi Cui, Wenxuan Qiu
Many Sparse Full State Reconstruction Sparse Optimization Regulatory Network Genome Understanding Inference Algorithm Gene Regulatory Neural Network

October 17, 2022

Principled Pruning of Bayesian Neural Networks through Variational Free Energy Minimization
Jim Beckers, Bart van Erp, Ziyue Zhao, Kirill Kondrashov, Bert de Vries
Bayesian Neural Network Free Energy Computational Neuroscience Model Reduction Inference Algorithm

October 13, 2022

Dirichlet process mixture models for non-stationary data streams
Ioar Casado, Aritz Pérez
Variational Inference Non Stationary Bayesian Nonparametric Inference Algorithm Streaming Inference Dirichlet Process Mixture Model

October 4, 2022

Amortized Bayesian Inference of GISAXS Data with Normalizing Flows
Maksim Zhdanov, Lisa Randolph, Thomas Kluge, Motoaki Nakatsutsumi, Christian Gutt, Marina Ganeva, Nico Hoffmann
Normalizing Flow Variational Auto Numerical Simulation Inference Algorithm Approximate Bayesian Computation Amortized Bayesian Inference

July 11, 2022

Synergy and Symmetry in Deep Learning: Interactions between the Data, Model, and Inference Algorithm
Lechao Xiao, Jeffrey Pennington
Deep Learning Convolutional Neural Network Full Model Data Dimensionality Cyclic Symmetry Model Synergy Malicious Node Fully Connected Inference Algorithm

June 24, 2022

Multi-Modal and Multi-Factor Branching Time Active Inference
Théophile Champion, Marek Grześ, Howard Bowman
Monte Carlo Tree Search Active Inference Inference Algorithm Likelihood Map

June 13, 2022

SIXO: Smoothing Inference with Twisted Objectives
Dieterich Lawson, Allan Raventós, Andrew Warrington, Scott Linderman
Scientific Inference Posterior Approximation Model Learning Sequential Monte Carlo Inference Algorithm

May 25, 2022

Unbiased and Efficient Sampling of Dependency Trees
Miloš Stanojević
Dependency Tree Efficient Sampling Physical Constraint Inference Algorithm

May 20, 2022

The price of ignorance: how much does it cost to forget noise structure in low-rank matrix estimation?
Jean Barbier, TianQi Hou, Marco Mondelli, Manuel Sáenz
Gaussian Noise Curious Price Inference Algorithm Low Rank Matrix Estimation Symmetric Noise

February 16, 2022

Analysis of Random Sequential Message Passing Algorithms for Approximate Inference
Burak Çakmak, Yue M. Lu, Manfred Opper
General Analysis Message Passing Approximate Inference Inference Algorithm Gaussian Latent Replica Theory