Skip to main content
Search

Seoyoon Koo

1 related articles

Subscribe to updates
  1. ADMET multitask grouping stops relying on trial and error: TAG-lite uses shared-encoder gradient affinity to split largely disjoint endpoints into seven groups

    ADMET endpoints are usually measured on different compounds, so existing task-compatibility measures that require shared compounds do not transfer directly. TAG-lite instead compares the direction of two tasks' gradients in the parameter space of a shared AttentiveFP encoder, estimating affinity from one multitask training configuration and clustering 27 TDC endpoints into seven groups. Affinity measured before group-wise training tracked group-level transfer (r = 0.881, permutation P = 0.006), but stopped tracking transfer magnitude below roughly 5% compound overlap.

You’re all caught up