The probabilistic approach has been responsible for most of the recent progress in artificial intelligence, such as voice recognition systems, or the system that recommends movies to Netflix subscribers. But Noah Goodman, an MIT research scientist whose department is Brain and Cognitive Sciences but whose lab is Computer Science and Artificial Intelligence, thinks that AI gave up too much when it gave up rules. By combining the old rule-based systems with insights from the new probabilistic systems, Goodman has found a way to model thought that could have broad implications for both AI and cognitive science.
That’s where probability processing shines, Ben Vigoda, Lyric Semiconductor’s CEO, told TechNewsWorld. It’s based on Pbits, or probability bits.
“Digital bits flow through Boolean logic gates, while Pbits flow through probability or Bayesian gates multi-directionally,” Vigoda said. “While digital processors program each operation in sequence, probability processors allow all the variables to talk to each other.”
A Pbit is the probability of a bit. Every event has two possibilities—it either happens or does not happen—and a Pbit encapsulates that, Vigoda said.
By letting all the variables talk to each other simultaneously, probability processors engage in both multidimensional and parallel processing, Vigoda pointed out.
Lyric’s chips are definitely a cool technology. But how do they compare with quantum computers? On the plus side, and it’s a very big plus, Lyric’s technology is already commercially available and very portable. QCs won’t be so for a long time. On the minus side, I suspect that D-Wave’s adiabatic quantum computer, and gate model quantum computers, once those are available for a large number of qubits, will be more efficient than Lyric’s chip when doing similar calculations.
I’d pay attention to this but note that the second source isn’t reliable. Anyone who is talking about D-Wave seriously doesn’t know much about quantum computing. Unfortunately, D-wave is a massively hyped project which has in practice done close to zero actual work. Scott Aaronson’s writing on the subject.
Probability & AI
A grand unified theory of AI
Now here is the hardware, New Chip Startup Plays the Odds on Probability Processing:
More, Bayes Chip:
I’d pay attention to this but note that the second source isn’t reliable. Anyone who is talking about D-Wave seriously doesn’t know much about quantum computing. Unfortunately, D-wave is a massively hyped project which has in practice done close to zero actual work. Scott Aaronson’s writing on the subject.