AI Paper Atlas

Research map / Probabilistic and scientific modeling / Dynamical systems

Adaptation Accelerating Sampling-based Bayesian Inference in Attractor Neural Networks

Xingsi Dong, Zilong Ji, Tianhao Chu, Tiejun Huang, Wenhao Zhang, Si Wu

NeurIPS 2022 · Probabilistic and scientific modeling › Dynamical systems · approach: Recurrent circuit models of inference · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add noisy adaptation (0 < m < mmax) to the CANN (ablation)CANN sampling of Gaussian posterior · single CANN, Fig 2B-Csampling convergence speed (KL to posterior): 2-4s -> 100-200msImproved
increase adaptation strength m beyond mmax toward mth (ablation)CANN with noisy adaptation · single CANN, Fig 2Esampling speedHurt
set adaptation strength above threshold mth (ablation)CANN with noisy adaptation · single CANNsampling validityHurt
vary input precision Lambda (ablation)CANN with noisy adaptation · single CANN, Fig 2FKL convergenceNo clear effect
add noisy adaptation to coupled CANNs (ablation)coupled CANNs, 2D features · Fig 3CKL convergence speedImproved
violate noise condition sigma_y^2 = 2 beta tau_s/tau_z (ablation)CANN with noisy adaptation · SI.3sampling performanceNo clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy