What is Cebra?
CEBRA, a cutting-edge machine-learning method, is transforming the field of neuroscience by revealing hidden structures in behavioral and neural data. Developed by a team at EPFL, this innovative tool excels in decoding activity from the mouse brain’s visual cortex, reconstructing viewed videos with remarkable accuracy. Its application ranges from rat hippocampus data to mouse primary visual cortex recordings, showcasing its versatility and precision in understanding neural dynamics during adaptive behaviors.
Key Features:
🧩 Joint Behavioral and Neural Analysis: CEBRA uniquely combines behavioral and neural data, offering a comprehensive understanding of the relationship between actions and neural activity.
🔄 Consistent and High-Performance Latent Spaces: By utilizing both supervised and self-supervised learning, CEBRA produces latent spaces that are both consistent and high in performance.
🔍 Decoding Capabilities: The inferred latents can be used for decoding, allowing for the reconstruction of videos from neural activity, a groundbreaking achievement in neuroscience.
🦎 Versatility Across Species and Behaviors: Validated on calcium and electrophysiology datasets, CEBRA is applicable to a wide range of sensory and motor tasks, across simple or complex behaviors in different species.
🔄 Label-Free Usage: CEBRA can be used in a label-free manner, making it a flexible tool for both hypothesis testing and discovery-driven research.





