Cebra

Introducing Cebra, a powerful machine learning tool that leverages non-linear techniques to create consistent and high-performance latent spaces from joint behavioural and neural data, making it a valuable asset for neuroscientists. This tool offers a range of features that allow for hypothesis testing and discovery-driven analysis, such as neural latent embeddings that are validated for accuracy with calcium and electrophysiology datasets and sensory and motor tasks.

Cebra is a versatile tool that can be used with single or multi-session datasets and without labels, providing high-accuracy decoding of natural movies from visual cortex. Its code is publicly available on GitHub, along with the pre-print on arxiv.org, for easy access and modification.

Some of Cebra's most notable use cases include the ability to analyze and decode behavioural and neural data to reveal underlying neural representations, map and uncover complex kinematic features in neuroscience research, and produce consistent latent spaces across various data types and experiments.

• Non-linear techniques used to create consistent and high-performance latent spaces from joint behavioural and neural data.
• Validated accuracy with calcium and electrophysiology datasets and sensory and motor tasks.
• Can be used with single or multi-session datasets and without labels.
• High-accuracy decoding of natural movies from visual cortex.
• Code is publicly available on GitHub and arxiv.org.
• Can analyze and decode behavioural and neural data, map complex kinematic features, and produce consistent latent spaces.

Cebra is the perfect tool for entrepreneurs in the neuroscience industry who are just beginning to explore the possibilities that AI-powered tools offer. Its easy-to-use features and the versatility make it ideal for anyone seeking to improve their work efficiency and profit.