AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

Learning Time-Invariant Representations for Individual Neurons from Population Dynamics

Lu Mi, Trung Le, Tianxing He, Eli Shlizerman, Uygar Sümbül

NeurIPS 2023 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Self-supervised neuroimaging representations · 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 permutation-invariant population summary inputs (ablation)NeuPRINT · Subclass prediction (MLP)top-1 accuracy: 0.512 -> 0.756Improved
add running speed input (ablation)NeuPRINT · Subclass prediction (MLP)top-1 accuracy: 0.512 -> 0.609Improved
add pupil size input (ablation)NeuPRINT · Subclass prediction (MLP)top-1 accuracy: 0.609 -> 0.658Improved
add frame state input (ablation)NeuPRINT · Subclass prediction (MLP)top-1 accuracy: 0.658 -> 0.683Improved
add population activity input (ablation)NeuPRINT · Subclass prediction (MLP)top-1 accuracy: 0.683 -> 0.732Improved
add center-surround activity input (ablation)NeuPRINT · Subclass prediction (MLP)top-1 accuracy: 0.732 -> 0.756Improved
add population inputs (ablation)NeuPRINT · Class prediction (MLP)top-1 accuracy: 0.752 -> 0.807No clear effect
replace linear dynamics model with transformer (ablation)NeuPRINT · Subclass prediction (MLP)top-1 accuracy: 0.512 -> 0.756Improved
replace RNN dynamics model with transformer (ablation)NeuPRINT · Subclass prediction (MLP)top-1 accuracy: 0.610 -> 0.756Improved
use MSE instead of Gaussian NLL objective (ablation)NeuPRINT transformer · Subclass prediction (MLP)top-1 accuracy: 0.707 -> 0.756Improved
use MSE instead of Gaussian NLL objective (ablation)NeuPRINT nonlinear model · Subclass prediction (MLP)top-1 accuracy: 0.537 -> 0.512Mixed
self-supervised NeuPRINT instead of supervised LOLCATNeuPRINT · Subclass prediction single mousetop-1 accuracy: 0.404 -> 0.537Improved

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