BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Neural population geometry and optimal coding of tasks with shared
  latent structure - Xiaolu Wang
DTSTART:20260220T140000Z
DTEND:20260220T150000Z
UID:TALK245062@talks.cam.ac.uk
CONTACT:Adam Triabhall
DESCRIPTION:This week we will discuss and debate a very recent preprint by
  Wakhloo and colleagues (2026).\n\nAbstract: “Animals can recognize late
 nt structures in their environment and apply this information to efficient
 ly navigate the world. Several works argue that the brain supports these a
 bilities by forming neural representations from which behaviorally relevan
 t variables can be read out across contexts and tasks. However\, it is unc
 lear which features of neural activity facilitate downstream readout. Here
  we analytically determine the geometric properties of neural activity tha
 t govern linear readout generalization on a set of tasks sharing a common 
 latent structure. We show that four statistics summarizing the dimensional
 ity\, factorization and correlation structures of neural activity determin
 e generalization. Early in learning\, optimal neural representations are l
 ower dimensional and exhibit higher correlations between single units and 
 task variables than late in learning. We support these predictions through
  biological and artificial neural data analysis. Our results tie the linea
 rly decodable information in neural population activity to its geometry”
  (Wakhloo et al.\, 2026).\n\nReference: Wakhloo\, A. J.\, Slatton\, W.\, &
  Chung\, S. (2026). Neural population geometry and optimal coding of tasks
  with shared latent structure. Nature Neuroscience. https://doi.org/10.103
 8/s41593-025-02183-y\n
LOCATION:https://cam-ac-uk.zoom.us/j/92612577704?pwd=MUtqMjVQdXNmUTVIYjRkM
 G1NUW9GZz09
END:VEVENT
END:VCALENDAR