Deep inference of latent dynamics with spatio-temporal super-resolution using selective backpropagation through time
This work addresses bandwidth limitations in neural interfaces for neuroscience research, offering potential power savings and improved data analysis, though it is incremental as it builds on existing sequential autoencoder frameworks.
The paper tackles the trade-off between spatial and temporal resolution in neural recordings by introducing selective backpropagation through time (SBTT), a training strategy that learns deep generative models of latent dynamics from data with varying observed variables, enabling spatio-temporal super-resolution. The result is more efficient and higher-fidelity characterization of neural population dynamics, with SBTT outperforming state-of-the-art methods in electrophysiology and calcium imaging applications.
Modern neural interfaces allow access to the activity of up to a million neurons within brain circuits. However, bandwidth limits often create a trade-off between greater spatial sampling (more channels or pixels) and the temporal frequency of sampling. Here we demonstrate that it is possible to obtain spatio-temporal super-resolution in neuronal time series by exploiting relationships among neurons, embedded in latent low-dimensional population dynamics. Our novel neural network training strategy, selective backpropagation through time (SBTT), enables learning of deep generative models of latent dynamics from data in which the set of observed variables changes at each time step. The resulting models are able to infer activity for missing samples by combining observations with learned latent dynamics. We test SBTT applied to sequential autoencoders and demonstrate more efficient and higher-fidelity characterization of neural population dynamics in electrophysiological and calcium imaging data. In electrophysiology, SBTT enables accurate inference of neuronal population dynamics with lower interface bandwidths, providing an avenue to significant power savings for implanted neuroelectronic interfaces. In applications to two-photon calcium imaging, SBTT accurately uncovers high-frequency temporal structure underlying neural population activity, substantially outperforming the current state-of-the-art. Finally, we demonstrate that performance could be further improved by using limited, high-bandwidth sampling to pretrain dynamics models, and then using SBTT to adapt these models for sparsely-sampled data.