Decoding the brain by training 1000 AI models to predict movement

Columbia University’s Paninski Lab developed Lightning Pose, a video-based, semi-supervised deep learning framework that improves animal movement tracking to support neuroscience research on the relationship between behavior and neural activity. Traditional model development would have required months to train and evaluate thousands of configurations, but using Lightning AI Studios for GPU resources, reproducible environments, and large-scale training sweeps allowed the team to launch 1,000 jobs in two weeks and reduce training time by 90%. Built with PyTorch Lightning, the system incorporates labeled frames, unlabeled videos, and motion-based constraints to produce more reliable predictions under conditions such as occlusions and changing camera angles. On mouse pupil-tracking data, it outperformed models such as DeepLabCut in prediction smoothness, pixel error, and correlations between tracked behavior and neural recordings. The platform also enabled easier sharing of complete, reproducible workspaces with collaborators and reviewers, and Lightning Pose is now used by groups including the International Brain Lab to analyze behavior across more than 30,000 experiments.

No tracked trend matches for this post yet.

Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.