CEBRA is an advanced machine-learning tool designed to analyze complex time series data, particularly focusing on neural and behavioral recordings. By compressing high-dimensional data, it uncovers hidden structures and patterns that traditional methods might overlook.
This capability is especially beneficial in neuroscience research, where understanding the relationship between behavior and neural activity is crucial.
Developed to handle both calcium imaging and electrophysiology datasets, CEBRA excels across various sensory and motor tasks. It can decode neural activity from the visual cortex to reconstruct viewed videos, demonstrating its high accuracy and utility.
Additionally, CEBRA supports both supervised and self-supervised analyses, making it adaptable for hypothesis-driven studies or exploratory data analysis.
