High-Performance Reservoir Computing With Fluctuations in Linear Networks
Nokkala, Johannes; Martínez-Peña, Rodrigo; Zambrini, Roberta; Soriano, Miguel C.
IEEE Transactions on Neural Networks and Learning Systems 33, 2664-2675 (2022)
IEEE Transactions on Neural Networks and Learning Systems 33, 2664-2675 (2022)
Reservoir computing has emerged as a powerful machine learning paradigm for harvesting nontrivial information processing out of disordered physical systems driven by sequential inputs. To this end, the system observables must become nonlinear functions of the input history. We show that encoding the input to quantum or classical fluctuations of a network of interacting harmonic oscillators can lead to a high performance comparable to that of a standard echo state network in several nonlinear benchmark tasks. This equivalence in performance holds even with a linear Hamiltonian and a readout linear in the system observables. Furthermore, we find that the performance of the network of harmonic oscillators in nonlinear tasks is robust to errors both in input and reservoir observables caused by external noise. For any reservoir computing system with a linear readout, the magnitude of trained weights can either amplify or suppress noise added to reservoir observables. We use this general result to explain why the oscillators are robust to noise and why having precise control over reservoir memory is important for noise robustness in general. Our results pave the way toward reservoir computing harnessing fluctuations in disordered linear systems.
Related research projects
QUAREC
Machine learning with quantum reservoir computing
I.P.: Roberta Zambrini
Project funded by the government of the Balearic Islands with the goal of extending Reservoir Computing into the quantum domain. The project is mainly theoretical/numerical, but also deals with the identification of …
QuaResC
Quantum machine learning using reservoir computing
I.P.: Miguel C. Soriano, Roberta Zambrini
The QuaResC project engages in a new collaboration UIB and CSIC researchers at IFISC with the objective to address an interdisciplinary topic between artificial intelligence and quantum physics: quantum machine learning using …
QuaResC (CSIC)
Quantum machine learning using reservoir computing
I.P.: Roberta Zambrini
The QuaResC project engages in a new collaboration UIB and CSIC researchers at IFISC with the objective to address an interdisciplinary topic between artificial intelligence and quantum physics: quantum machine learning using …
QuaResC (UIB)
Quantum machine learning using reservoir computing
I.P.: Miguel C. Soriano
The QuaResC project engages in a new collaboration UIB and CSIC researchers at IFISC with the objective to address an interdisciplinary topic between artificial intelligence and quantum physics: quantum machine learning using …