Probability-processing hardware

Lyric Semiconductor posted:

For over 60 years, computers have been based on digital computing principles. Data is represented as bits (0s and 1s). Boolean logic gates perform operations on these bits. A processor steps through many of these operations serially in order to perform a function. However, today’s most interesting problems are not at all suited to this approach.

Here at Lyric Semiconductor, we are redesigning information processing circuits from the ground up to natively process probabilities: from the gate circuits to the processor architecture to the programming language. As a result, many applications that today require a thousand conventional processors will soon run in just one Lyric processor, providing 1,000x efficiencies in cost, power, and size.

Om Malik has some more information, also relating to the team and the business.

The fundamental idea is that computing architectures work deterministically, even though the world is fundamentally stochastic. In a lot of statistical processing, especially in Bayesian statistics, we take stochastic world, force it into determinism, simulate stochastic world by computationally generating deterministic pseudo-random numbers, and simulate stochastic matching by deterministic likelihood computations. What Lyric could do is to bypass this highly inefficient intermediate deterministic step. This way, we’ll be able to fit bigger and better models much faster.

They’re also working on a programming language: PSBL (Probability Synthesis for Bayesian Logic), but there are no details. Here is their patent for Analog Logic Automata, indicating applications for images (filtering, recognition, etc.).

[D+1: J.J. Hayes points to a US Patent indicating that one of the circuits optimizes the sum-product belief propagation algorithm. This type of algorithms is popular in machine learning for various recognition and denoising problems. One way to explain it to a statistician is that it does imputation with an underlying loglinear model.]