DOI,IST REx ID,Research Group,Title of publication
null,14179,FrLo,Self-supervised learning with data augmentations provably isolates content from style
null,14180,FrLo,Dynamic inference with neural interpreters
null,14182,FrLo,Backward-compatible prediction updates: A probabilistic approach
null,10593,MaMo,PCA initialization for approximate message passing in rotationally invariant models
null,10594,MaMo,When are solutions connected in deep networks?
null,11452,DaAl,Distributed principal component analysis with limited communication
null,11453,TiVo,Online learning of neural computations from sparse temporal feedback
null,11458,"GradSch,DaAl",AC/DC: Alternating Compressed/DeCompressed training of deep neural networks
null,11463,DaAl,M-FAC: Efficient matrix-free approximations of second-order information
null,11464,DaAl,Towards tight communication lower bounds for distributed optimisation