Are AI Models Working Harder Than They Need to?
Much of modern AI runs on multiplication. Neural networks behind everything from generated answers to photo organization and song recommendations perform millions or billions of operations that multiply inputs by learned weights. John thinks that's more work than the job requires.
Key Takeaways
- Instead of repeatedly multiplying inputs by weights, these networks pass binary inputs through interconnected lookup tables-closer to consulting a collection of stored answers than solving the same arithmetic problem repeatedly.
Depending on the task, she says, the networks can be less than a thousandth the size or 1,000 times as fast as conventional alternatives while maintaining comparable accuracy.
- Lima at the Federal University of Rio de Janiero kept working on it for years, just one research group, so not as much work as in the conventional neural networks.
My friend knew I had hardware implementation experience, so he thought I could help make it real.
- In less than six months, we had it running on an FPGA, basically a ready-made chip.
We were able to create some very small neural networks that used the lookup methodology that could fit on tiny chips and didn't need a GPU to run them.
- I'm thinking, yes, no, I should say this, I shouldn't say that.
And the human brain only consumes about 20 watts of energy doing that.
- As children, we all struggled to learn the multiplication tables.
Stats & Key Facts
- #Depending on the task, she says, the networks can be less than a thousandth the size or 1,000 times as fast as conventional alternatives while maintaining comparable accuracy.
- #We could get them 1,000 times smaller than what everyone else was getting.
- #The model behind most popular networks today is based on a neuron model from a 1943 paper, the McCulloch-Pitts model .
John, a professor of electrical and computer engineering at the University of Texas at Austin, has spent the past five years working on a class of models called weightless neural networks . Instead of repeatedly multiplying inputs by weights, these networks pass binary inputs through interconnected lookup tables-closer to consulting a collection of stored answers than solving the same arithmetic problem repeatedly. Depending on the task, she says, the networks can be less than a thousandth the size or 1,000 times as fast as conventional alternatives while maintaining comparable accuracy.
Her team's work has so far focused on small, specific problems: medical sensors, activity tracking, keyword spotting. But she thinks the same approach could eventually reach much bigger targets, including the transformer models behind today's chatbots. What made you walk away from weights in the first place?
Was there a specific moment that pushed you toward lookups instead? John: A friend casually invited me to a weekly meeting a few years ago to talk about using lookups instead of weights, a technique that wasn't new. Someone in the U.K. had built a commercial product around it in the '80s for pattern recognition, and then it just disappeared.
For more details please read the original article at IEEE Spectrum AI.
Continue Learning
Comments
Comments appear only after moderation. Your email identifies your submission to the moderator and is never displayed here.
No approved comments yet.