Energetics of computation in artificial and natural networks
Thursday – Salle Dussane, Ecole Normale Supérieure, 45 rue d’Ulm, Paris 5th
Friday – Site Ulm, Collège de France, 3 rue d’Ulm, Paris 5th
While improving performance is a top priority in computation and machine learning the issue of energy consumption is a matter of increasing interest, either in specific situations (such as portable devices) or due to general concerns about sustainability. Computing with limited resources has always been a fundamental constraint for living organisms, and various strategies have been discovered in the course of Evolution, implementing trade-offs between speed, accuracy and energy consumption. Biology can also be a source of inspiration to develop new computing devices, together with progress in new materials, e,g, spintronics, photonics, and others. The aim of this workshop is to gather experts with diverse backgrounds, ranging from neuroscience to machine learning, from material sciences and engineering to theory, all interested in the energetics of computation.