Energy Complexity of Recurrent Neural Networks

0393985 - ÚI 2015 RIV US eng J - Journal Article
Šíma, Jiří
Energy Complexity of Recurrent Neural Networks.
Neural Computation. Roč. 26, č. 5 (2014), s. 953-973. ISSN 0899-7667. E-ISSN 1530-888X
R&D Projects: GA ČR GAP202/10/1333
Institutional support: RVO:67985807
Keywords : neural network * finite automaton * energy complexity * optimal size
Subject RIV: IN - Informatics, Computer Science
Impact factor: 2.207, year: 2014

Recently, a new so-called energy complexity measure has been introduced and studied for feedforward perceptron networks. This measure is inspired by the fact that biological neurons require more energy to transmit a spike than not to fire, and the activity of neurons in the brain is quite sparse, with only about 1% of neurons firing. In this paper, we investigate the energy complexity of recurrent networks which counts the number of active neurons at any time instant of a computation. We prove that any deterministic finite automaton with m states can be simulated by a neural network of optimal size s=\Theta(\sqrt{m}) with the time overhead of \tau=O(s/e) per one input bit, using the energy O(e), for any e such that e=\Omega(\log s) and e=O(s), which shows the time-energy tradeoff in recurrent networks. In addition, for the time overhead \tau satisfying \tau^\tau=o(s), we obtain the lower bound of s^{c/\tau} on the energy of such a simulation, for some constant c>0 and for infinitely many s.
Permanent Link: http://hdl.handle.net/11104/0222343