Join Zoom Meeting https://zoom.us/j/475819702
Meeting ID: 475-819-702

In this talk, I will describe recent work on analyzing behavioral dynamics using the theory of nonlinear dynamics. I will first describe ways to convert time-series observations into a phase-space, which is a maximally predictive geometric representation of the data. The phase-space, once properly reconstructed, contains all dynamically relevant information available in the measurements. Next, I will describe methods to extract meaningful information from the phase space by analyzing its local geometry and topology. Applications include separating continuous behaviors into discrete motifs, identifying salient events, extracting long time-scale structure, quantifying behavioral variability and studying coupling between sub-systems.

 

 

References:

 

1. Ahamed, Tosif, Antonio C. Costa, and Greg J. Stephens. "Capturing the continuous complexity of behaviour in Caenorhabditis elegans." Nature Physics 17.2 (2021): 275-283.

2. Costa, Antonio Carlos, et al. "Maximally predictive ensemble dynamics from data." arXiv preprint arXiv:2105.12811 (2021).