Search papers, labs, and topics across Lattice.
University of Southern California
3
0
4
TabPFN slashes negative log-likelihood by up to 62% and drastically improves calibration in multimodal classification without requiring any training.
Norm-constrained deep networks can dodge the curse of dimensionality when learning compositional functions, unlocking efficient learning even in overparameterized regimes.
Sparsity could be the key to unlocking efficient neural networks for learning operators on infinite-dimensional function spaces, sidestepping the curse of dimensionality.
