-
. Going beyond the canonical sub-Gaussian noise models, the objective is to prove tight convergence rates for first-order or zeroth-order methods when the noise is heavy tailed. This allows us to reliably
-
prototyping, fabrication experiments, development of experimental setups, material testing, design modelling and optimisation, and the preparation of workflows interfacing with robotic and construction
-
, circuit discovery, activation patching, and representation engineering, with a focus on compositional structure in learned representations, as well as testing universality across models and languages
Searches related to testing
Enter an email to receive alerts for testing positions