Representation learning for neural population activity with Neural Data Transformers
This work addresses the need for faster neural data modeling in real-time applications like brain-computer interfaces, though it is incremental as it builds on existing transformer architectures.
The paper tackled the problem of slow sequential processing in modeling neural population dynamics for real-time applications by introducing the Neural Data Transformer (NDT), a non-recurrent model that matches state-of-the-art recurrent models in accuracy and achieves 3.9ms inference, over 6 times faster than baselines.
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