We develop a general duality between neural networks and compositional kernel Hilbert spaces. We introduce the notion of a computation skeleton, an acyclic graph that succinctly describes both a family of neural networks and a kernel space. Random neural networks are generated from a skeleton through node replication followed by sampling from a normal distribution to assign weights. The kernel space consists of functions that arise by compositions, averaging, and non-linear transformations governed by the skeleton's graph topology and activation functions. We prove that random networks induce representations which approximate the kernel space. In particular, it follows that random weight initialization often yields a favorable starting point for optimization despite the worst-case intractability of training neural networks.
|Original language||American English|
|Number of pages||9|
|Journal||Advances in Neural Information Processing Systems|
|State||Published - 2016|
|Event||30th Annual Conference on Neural Information Processing Systems, NIPS 2016 - Barcelona, Spain|
Duration: 5 Dec 2016 → 10 Dec 2016
Bibliographical notePublisher Copyright:
© 2016 NIPS Foundation - All Rights Reserved.