July 1 h. 9:00-13:00 lab 146
July 3 h. 9:00-13:00 15:00-17:00 lab 146
July 4 h. 9:00-13:00 15:00-17:00 lab 146
July 5 h. 9:00-13:00 lab 146
Techniques and technologies for the design of memristor-based nonlinear circuits as fundamental building-block of advanced cognitive intelligent systems.
Knowledge of basic circuit theory and of electronic devices. Basic principles of nonlinear dynamics.
Knowledge: basics of physics for circuit modeling, analysis and design, in particular nonlinear circuit theory. Knowledge of nonlinear differential ordinary equations and nonlinear analysis.
Skills: modelling of nanoelectronic devices and circuit analysis. Qualitative analysis on nonlinear dynamical systems (equilibrium points, periodic/chaotic attractors and bifurcation phenomena).
Competences: students will reach a sufficient knowledge and skill for being able to analyze and design nonlinear circuits with memristors. In addition, students will acquire the fundamental principles of neural networks and machine learning algorithms.
After the computer and internet revolutions in the last 50 years, internet of things (IoT) promises to be the next big technology revolution deeply impacting all aspects of human life, e.g., industrial processes, health, transport, communication, and many others. The retrieve of relevant information from massive amount of data will soon be impossible with conventional computers due to physical limitations.
A major challenge is to bring sufficient intelligence on board of the system, while maintaining reasonably low power consumption and real time reconfiguration. Such objective might be achieved via memristive hardware, which has recently raised increasing interest as low power, high-density systems for learning and recognition of patterns, such as handwritten characters, faces, and speech. In this course I will present work that has been done towards the development of memristor-based cognitive systems. In particular the talk covers the fundamental theory of memristor circuits and their nonlinear dynamical properties (e.g. synchronization, spatio-temporal pattern,...), the spiking computing principles with memristor synapses and the use of memristor networks in pattern recognition tasks.
Overview of existing memristor technology
Background and Challenges
Memristor Devices and Models
Fundamental Theory of Memristor
Flux-Charge Analysis Method
Nonlinear Dynamics in Memristor Circuits
Synchronization and Oscillatory Associative Memories
Neuromorphic Computer Architectures
Cellular Nanoscale/Nonlinear Networks with Memristors