2019 RecentAdvancesinPhysicalReservo

• (Tanaka et al., 2019) ⇒ Gouhei Tanaka, Toshiyuki Yamane, Jean Benoit Héroux, Ryosho Nakane, Naoki Kanazawa, Seiji Takeda, Hidetoshi Numata, Daiju Nakano, and Akira Hirose. (2019). “Recent Advances in Physical Reservoir Computing: A Review.” In: Neural Networks - Elsevier Journal, 115. doi:10.1016/j.neunet.2019.03.005

Subject Headings: Reservoir Computing

Notes

• ArXiv: 1808.04962

Cited By

• Google Scholar: ~ 22 Citations (Retrieved:2017-10-27).
• Semantic Scholar: ~ 12 Citations (Retrieved:2017-10-27).
• MS Academic: ~ 21 Citations (Retrieved:2017-10-27).

Quotes

Author Keywords

• Neural networks; Machine learning; Reservoir computing; Nonlinear dynamical systems; Neuromorphic device

Abstract

Reservoir computing is a computational framework suited for temporal / sequential data processing. It is derived from several recurrent neural network models, including echo state networks and liquid state machines. A reservoir computing system consists of a reservoir for mapping inputs into a high-dimensional space and a readout for pattern analysis from the high-dimensional states in the reservoir. The reservoir is fixed and only the readout is trained with a simple method such as linear regression and classification. Thus, the major advantage of reservoir computing compared to other recurrent neural networks is fast learning, resulting in low training cost. Another advantage is that the reservoir without adaptive updating is amenable to hardware implementation using a variety of physical systems, substrates, and devices. In fact, such physical reservoir computing has attracted increasing attention in diverse fields of research. The purpose of this review is to provide an overview of recent advances in physical reservoir computing by classifying them according to the type of the reservoir. We discuss the current issues and perspectives related to physical reservoir computing, in order to further expand its practical applications and develop next-generation machine learning systems.

3. Dynamical systems models for RC

(...)

(...) As shown in Fig. 3, the input signal is time-multiplexed by a mask function (Appeltant, Van der Sande, Danckaert, & Fischer, 2014) and fed to the single nonlinear node. The virtual nodes are set at time points that equally divide the delay period . The time interval between two consecutive nodes is . The states at these virtual nodes, for , are used as the reservoir state at time and then fed to the output layer through weighted connections. These connection weights are trained in the readout. The system was successfully applied to the spoken digit recognition task and the nonlinear autoregressive moving average (NARMA)-10 time series prediction task.

The architecture of the single-node reservoir with delayed feedback was extended in two ways (Ortín & Pesquera, 2017). One is an ensemble of two separate time-delayed reservoirs whose outputs are combined at the readout. The other is a circular concatenation of the delay lines of two reservoirs, forming a longer delay line. These extended architectures were shown to achieve better performance, faster processing speed, and higher robustness than the single-node reservoir. An extensive amount of work has been performed on single-node reservoirs with delayed feedback (Brunner et al., 2018).

The simplicity of the single-node reservoir with delayed feedback is advantageous for physical implementation compared with network-based reservoirs consisting of a large number of nodes. In fact, single-node reservoirs have been widely employed for electronic RC (Section 4) and photonic RC (Section 5.2).

Figure 3. RC using a single nonlinear node reservoir with time-delayed feedback (Appeltant et al., 2011).

References

;

	Author	volume	Date Value	title	type	journal	titleUrl	doi	note	year
2019 RecentAdvancesinPhysicalReservo	Gouhei Tanaka Toshiyuki Yamane Jean Benoit Héroux Ryosho Nakane Naoki Kanazawa Seiji Takeda Hidetoshi Numata Daiju Nakano Akira Hirose			Recent Advances in Physical Reservoir Computing: A Review				10.1016/j.neunet.2019.03.005		2019