The mise en scene of memristive networks: effective memory, dynamics and learning
arXiv:1611.02104 · doi:10.1080/17445760.2017.1320796
Abstract
We discuss the properties of the dynamics of purely memristive circuits using a recently derived consistent equation for the internal memory variables of the involved memristors. In particular, we show that the number of independent memory states in a memristive circuit is constrained by the circuit conservation laws, and that the dynamics preserves these symmetries by means of a projection on the physical subspace. Moreover, we discuss other symmetries of the dynamics under various transformations of the internal memory, and study the linearized and strongly non-linear regimes of the dynamics. In the strongly non-linear regime, we derive a conservation law for the internal memory variables. We also provide a condition on the reality of the eigenvalues of Lyapunov matrices describing the linearized dynamics close to a fixed point. We show that the eigenvalues ca be imaginary only for mixtures of passive and active components. Our last result concerns the weak non-linear regime. We show that the internal memory dynamics can be interpreted as a constrained gradient descent, and provide the functional being minimized. This latter result provides another direct connection between memristors and learning.
arXiv comment19 pages, 2 figures; more typos corrected, proof of fixed point corrected, presentation improved
References in corpus (18)
- Emergent complex neural dynamics
- Circuit elements with memory: memristors, memcapacitors and meminductors
- Memory and information processing in neuromorphic systems
- Memory effects in complex materials and nanoscale systems
- Memristive model of amoeba's learning
- Memcomputing: a computing paradigm to store and process information on the same physical platform
- Solving mazes with memristors: a massively-parallel approach
- Universal Memcomputing Machines
- On the physical properties of memristive, memcapacitive, and meminductive systems
- Quantum Memristors
- Polynomial-time solution of prime factorization and NP-hard problems with digital memcomputing machines
- Memcomputing NP-complete problems in polynomial time using polynomial resources and collective states
- Self-organization and solution of shortest-path optimization problems with memristive networks
- Memristor models for machine learning
- Memory models of adaptive behaviour
- The complex dynamics of memristive circuits: analytical results and universal slow relaxation
- Scale-free networks as an epiphenomenon of memory
- Cycle/cocycle oblique projections on oriented graphs