Get the latest tech news
Analog optical computer for AI inference and combinatorial optimization
An analog optical computer that combines analog electronics, three-dimensional optics, and an iterative architecture accelerates artificial intelligence inference and combinatorial optimization in a single platform, paving a promising path for faster and sustainable computing.
To illustrate its versatility, the remainder of the paper presents four case studies highlighting how equilibrium ML models can be applied to classification and regression tasks, and how the QUMO paradigm can represent real-world applications in finance and healthcare, while utilizing the same AOC hardware. The current AOC hardware uses a rapid fixed-point search to power inference tasks, such as regression and classification, using equilibrium models with promising reasoning capabilities, and to successfully solve QUMO problems including medical image reconstruction and transaction settlement. Kirill P. Kalinin, Jannes Gladrow, Jiaqi Chu, James H. Clegg, Daniel Cletheroe, Douglas J. Kelly, Babak Rahmani, Grace Brennan, Burcu Canakci, Fabian Falck, Heiner Kremer, Greg O’Shea, Lucinda Pickup, Ant Rowstron, Christos Gkantsidis, Francesca Parmigiani & Hitesh Ballani
Or read this on Hacker News