Neuromorphic Computing: From Materials to Systems Architecture - Report of a Roundtable Convened to Consider Neuromorphic Computing Basic Research Needs


30 Mar 2016 Gaithersburg - The report of a Roundtable Convened to Consider Neuromorphic Computing Basic Research Needs at the Neuromorphic Computing: From Materials to Systems Architecture in Gaithersburg, October 2015, has been published. The basic question was: "Can brain-like (neuromorphic) computing devices based on new material concepts and systems be developed to dramatically outperform conventional CMOS based technology? If so, what are the basic research challenges for materials science and computing?"

The main conclusions of the roundtable were:

  1. Creating the architectural design for neuromorphic computing requires an integrative, interdisciplinary approach between computer scientists, engineers, physicists, and materials scientists
  2. Creating a new computational system will require developing new system architectures to accommodate all needed functionalities
  3. One or more reference architectures should be used to enable comparisons of alternative devices and materials
  4. The basis for the devices to be used in these new computational systems require the development of novel nano and meso structured materials; this will be Neuromorphic Computing: From Materials to Systems Architecture accomplished by unlocking the properties of quantum materials based on new materials physics
  5. The most promising materials require fundamental understanding of strongly correlated materials, understanding formation and migration of ions, defects and clusters, developing novel spin based devices, and/or discovering new quantum functional materials
  6. To fully realize open opportunities requires designing systems and materials that exhibit self- and external-healing, three-­dimensional reconstruction, distributed power delivery, fault tolerance, co-location of memory and processors, multistate, i.e., systems in which the present response depends on past history and multiple interacting state variables that define the present state
  7. The development of a new brainlike computational system will not evolve in a single step; it is important to implement well-­defined intermediate steps that give useful scientific and technological information

Ad Emmen