Goto

Collaborating Authors

 Africa



HardnessofNoise-FreeLearningfor Two-Hidden-LayerNeuralNetworks

Neural Information Processing Systems

We give superpolynomial statistical query (SQ) lower bounds for learning twohidden-layer ReLU networks with respect to Gaussian inputs in the standard (noise-free)model.






178b6cd141e003fa5ff808c7d7d8e2cc-Paper-Conference.pdf

Neural Information Processing Systems

AStackelberg game[30, 31] is a strategic interaction between two utility-maximizing players in which one player (theleader) is able to commit to a (possibly mixed) strategy before the other player (thefollower)takesanaction.