Charles Boutens Computing Optimal Input Encoding for Memristor Based Reservoir
Total Page:16
File Type:pdf, Size:1020Kb
Load more
Recommended publications
-
Reservoir Transformers
Reservoir Transformers Sheng Sheny, Alexei Baevskiz, Ari S. Morcosz, Kurt Keutzery, Michael Auliz, Douwe Kielaz yUC Berkeley; zFacebook AI Research [email protected], [email protected] Abstract is more, we find that freezing layers may actually improve performance. We demonstrate that transformers obtain im- Beyond desirable efficiency gains, random lay- pressive performance even when some of the ers are interesting for several additional reasons. layers are randomly initialized and never up- dated. Inspired by old and well-established Fixed randomly initialized networks (Gallicchio ideas in machine learning, we explore a variety and Scardapane, 2020) converge to Gaussian pro- of non-linear “reservoir” layers interspersed cesses in the limit of infinite width (Daniely et al., with regular transformer layers, and show im- 2016), have intriguing interpretations in metric provements in wall-clock compute time until learning (Rosenfeld and Tsotsos, 2019; Giryes convergence, as well as overall performance, et al., 2016), and have been shown to provide on various machine translation and (masked) excellent “priors” either for subsequent learn- language modelling tasks. ing (Ulyanov et al., 2018) or pruning (Frankle and 1 Introduction Carbin, 2018). Fixed layers allow for efficient low-cost hardware implementations (Schrauwen Transformers (Vaswani et al., 2017) have dom- et al., 2007) and can be characterized using only a inated natural language processing (NLP) in re- random number generator and its seed. This could cent years, from large scale machine transla- facilitate distributed training and enables highly tion (Ott et al., 2018) to pre-trained (masked) efficient deployment to edge devices, since it only language modeling (Devlin et al., 2018; Rad- requires transmission of a single number. -
The Promise of Spintronics for Unconventional Computing
The promise of spintronics for unconventional computing Giovanni Finocchio,1,* Massimiliano Di Ventra,2 Kerem Y. Camsari,3 Karin Everschor‐Sitte,4 Pedram Khalili Amiri,5 and Zhongming Zeng6 1Department of Mathematical and Computer Sciences, Physical Sciences and Earth Sciences, University of Messina, Messina, 98166, Italy ‐ Tel. +39.090.3975555 2Department of Physics, University of California, San Diego, La Jolla, California 92093, USA, ‐ Tel. +1.858.8226447 3School of Electrical and Computer Engineering, Purdue University, West Lafayette, Indiana 47907, USA ‐ Tel. +1.765.4946442 4Institut für Physik, Johannes Gutenberg‐Universität Mainz, DE‐55099 Mainz, Germany ‐ Tel. +49.61313923645 5Department of Electrical and Computer Engineering, Northwestern University, Evanston, Illinois 60208, USA ‐ Tel. +1.847.4671035 6 Key Laboratory of Multifunctional Nanomaterials and Smart Systems,, Suzhou Institute of Nano‐tech and Nano‐bionics, Chinese Academy of Sciences, Ruoshui Road 398, Suzhou 215123, P. R. China ‐ Tel. +86.0512.62872759 1 Abstract Novel computational paradigms may provide the blueprint to help solving the time and energy limitations that we face with our modern computers, and provide solutions to complex problems more efficiently (with reduced time, power consumption and/or less device footprint) than is currently possible with standard approaches. Spintronics offers a promising basis for the development of efficient devices and unconventional operations for at least three main reasons: (i) the low-power requirements of spin-based devices, i.e., requiring no standby power for operation and the possibility to write information with small dynamic energy dissipation, (ii) the strong nonlinearity, time nonlocality, and/or stochasticity that spintronic devices can exhibit, and (iii) their compatibility with CMOS logic manufacturing processes. -
Quantum Reservoir Computing: a Reservoir Approach Toward Quantum Machine Learning on Near-Term Quantum Devices
Quantum reservoir computing: a reservoir approach toward quantum machine learning on near-term quantum devices Keisuke Fujii1, ∗ and Kohei Nakajima2, y 1Graduate School of Engineering Science, Osaka University, 1-3 Machikaneyama, Toyonaka, Osaka 560-8531, Japan. 2Graduate School of Information Science and Technology, The University of Tokyo, Bunkyo-ku, 113-8656 Tokyo, Japan (Dated: November 11, 2020) Quantum systems have an exponentially large degree of freedom in the number of particles and hence provide a rich dynamics that could not be simulated on conventional computers. Quantum reservoir computing is an approach to use such a complex and rich dynamics on the quantum systems as it is for temporal machine learning. In this chapter, we explain quantum reservoir computing and related approaches, quantum extreme learning machine and quantum circuit learning, starting from a pedagogical introduction to quantum mechanics and machine learning. All these quantum machine learning approaches are experimentally feasible and effective on the state-of-the-art quantum devices. I. INTRODUCTION a useful task like machine learning, now applications of such a near-term quantum device for useful tasks includ- Over the past several decades, we have enjoyed expo- ing machine leanring has been widely explored. On the nential growth of computational power, namely, Moore's other hand, quantum simulators are thought to be much law. Nowadays even smart phone or tablet PC is much easier to implement than a full-fledged universal quan- more powerful than super computers in 1980s. Even tum computer. In this regard, existing quantum sim- though, people are still seeking more computational ulators have already shed new light on the physics of power, especially for artificial intelligence (machine learn- complex many-body quantum systems [9{11], and a re- ing), chemical and material simulations, and forecasting stricted class of quantum dynamics, known as adiabatic complex phenomena like economics, weather and climate. -
Arxiv:1808.04962V3 [Cs.ET] 15 Apr 2019 Dynamical Systems, Neuromorphic Device 2010 MSC: 68Txx, 37N20
Recent Advances in Physical Reservoir Computing: A Review Gouhei Tanakaa;b;1, Toshiyuki Yamanec, Jean Benoit H´erouxc, Ryosho Nakanea;b, Naoki Kanazawac, Seiji Takedac, Hidetoshi Numatac, Daiju Nakanoc, and Akira Hirosea;b aInstitute for Innovation in International Engineering Education, Graduate School of Engineering, The University of Tokyo, Tokyo 113-8656, Japan bDepartment of Electrical Engineering and Information Systems, Graduate School of Engineering, The University of Tokyo, Tokyo 113-8656, Japan cIBM Research { Tokyo, Kanagawa 212-0032, Japan 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 comput- ing 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 advan- tage of reservoir computing compared to other recurrent neural networks is fast learning, resulting in low training cost. Another advantage is that the reser- voir 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 re- search. 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.