Tensor networks in machine learning

A tensor network is a type of decomposition used to express and approximate large arrays data. given data-set, quantum state or higher dimensional multi-linear map factored approximated by composition smaller maps. This reminiscent how Boolean function might be decomposed into gate array: this represents special case decomposition, in which the entries are replaced 0, 1 factorisation becomes exact. The collection associated techniques called, methods: subject developed independently several distinct fields study, have more recently become interrelated through language networks. tantamount questions field relate expressability networks reduction computational overheads. merger with machine learning natural. On one hand, can aid determining factorization approximating data set. other structure viewed as model. Herein parameters adjusted learn classify data-set. In survey we recover basics explain ongoing effort develop theory learning.