Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Representation learning for neural population activity with Neural Data Transformers

About

Neural population activity is theorized to reflect an underlying dynamical structure. This structure can be accurately captured using state space models with explicit dynamics, such as those based on recurrent neural networks (RNNs). However, using recurrence to explicitly model dynamics necessitates sequential processing of data, slowing real-time applications such as brain-computer interfaces. Here we introduce the Neural Data Transformer (NDT), a non-recurrent alternative. We test the NDT's ability to capture autonomous dynamical systems by applying it to synthetic datasets with known dynamics and data from monkey motor cortex during a reaching task well-modeled by RNNs. The NDT models these datasets as well as state-of-the-art recurrent models. Further, its non-recurrence enables 3.9ms inference, well within the loop time of real-time applications and more than 6 times faster than recurrent baselines on the monkey reaching dataset. These results suggest that an explicit dynamics model is not necessary to model autonomous neural population dynamics. Code: https://github.com/snel-repo/neural-data-transformers

Joel Ye, Chethan Pandarinath• 2021

Related benchmarks

TaskDatasetResultRank
Neural Latents Benchmark evaluationMC_Maze Neural Latents Benchmark (NLB) (test)
co-bps0.3597
12
RegressionArea2-Bump (New Region)
R2 Score0.902
10
ClassificationPPC-FINGER New Specie
Balanced Accuracy (B-Acc)95.9
10
RegressionPerich T-RT New Task
R2 Score0.687
10
RegressionPerich T-CO New Subject
R2 Score73.7
10
RegressionMC-Maze
R2 Score0.882
10
ClassificationM1-CO1 (Cross-day)
B-Acc81.6
10
ClassificationLICK (New Subject)
Balanced Accuracy72.1
10
ClassificationM1-CO1 (Multi-day)
Balanced Accuracy75.5
10
ClassificationLICK 80% (test)
Balanced Accuracy66
9
Showing 10 of 20 rows

Other info

Follow for update