09 August 2014

A brain on a chip?

In Star Trek: Voyager, the titular spaceship had computers that ran on "bio-neural circuitry" stored in gel packs. Like human bodies, the bio-neural circuity was prone to viral infection and, in one episode, is treated with a makeshift "fever" created by an "inverted warp field", because Star Trek.

How far-fetched is the idea? As it turns out, not very. Scientists at IBM have developed a chip that mimics the neural structure of a brain. Wired explains:
In a [conventional] von Neumann computer, the storage and handling of data is divvied up between the machine’s main memory and its central processing unit. To do their work, computers carry out a set of instructions, or programs, sequentially by shuttling data from memory (where it’s stored) to the CPU (where it’s crunched). Because the memory and CPU are separated, data needs to be transferred constantly.
[....]
Neuromorphic chips developed by IBM and a handful of others don’t separate the data-storage and data-crunching parts of the computer. Instead, they pack the memory, computation and communication parts into little modules that process information locally but can communicate with each other easily and quickly. This, IBM researchers say, resembles the circuits found in the brain, where the separation of computation and storage isn’t as cut and dry, and it’s what buys the thing added energy efficiency—arguably the chip’s best selling point to date.
It's an interesting concept, and as the New York Times notes, it is both power-efficient and capable of massive parallel processing:
The chip contains 5.4 billion transistors, yet draws just 70 milliwatts of power. By contrast, modern Intel processors in today’s personal computers and data centers may have 1.4 billion transistors and consume far more power — 35 to 140 watts.
Today’s conventional microprocessors and graphics processors are capable of performing billions of mathematical operations a second, yet the new chip system clock makes its calculations barely a thousand times a second. But because of the vast number of circuits working in parallel, it is still capable of performing 46 billion operations a second per watt of energy consumed, according to IBM researchers.
Large-scale applications are still a long ways off, and unlike brains and conventional computers, the new chips can't learn. So maybe we're a ways off from supercomputers bearing any resemblance to the human brain. And these new chips are a good many steps away from the Voyager computers because they just mimic an aspect of brain structure — they don't actually contain organic matter.

Still, it's kinda cool to think about. Yann LeCun, a researcher in the field, is skeptical:
“This avenue of research is not going to pan out for quite a while, if ever. They may get neural net accelerator chips in their smartphones soonish, but these chips won’t look at all like the IBM chip. They will look more like modified GPUs.”
So, like, Assassin's Creed Unity in 4k resolution on my gaming PC? I guess that'll hold me over until we build some starships.

No comments:

Post a Comment

Note: Only a member of this blog may post a comment.