Country
Private Estimationwith Public Data
Lemma 2.2(zCDPPrivate Gaussian Mean Estimation).
CaMP: Causal Multi-policy Planning for Interactive Navigation in Multi-room Scenes Xiaohan Wang
Visual navigation has been widely studied under the assumption that there may be several clear routes to reach the goal.
FormalizingtheGeneralization-ForgettingTrade-Off inContinualLearning
In continual learning (CL), we incrementally adapt a model to learn tasks (defined according to the problem at hand) observed sequentially. CL has two main objectives: maintain long-term memory (remember previous tasks) and navigate new experiences continually (quickly adapt to newtasks).
Provably Strict Generalisation Benefit for Invariance in Kernel Methods
It is a commonly held belief that enforcing invariance improves generalisation.