Returning nearer to the current day, we discover industrial growth of AI beholden to “The Bitter Lesson.” After Nvidia’s CUDA enabled environment friendly tensor operations on GPUs and deep networks like AlexNet drove unprecedented progress in different fields, the beforehand numerous strategies competing for dominance in machine studying benchmarks homogenized to solely throwing extra compute at deep studying.
There’s maybe no better instance of the bitter lesson than massive language fashions, which displayed unimaginable emergent capabilities with scaling over the previous decade. Might we actually attain synthetic common intelligence (AGI), that’s, methods amounting to the archetypal depictions of AI seen in Blade Runner or 2001: A House Odyssey, just by including extra parameters to those LLMs and extra GPUs to the clusters they’re skilled on?
My work at UCSD was predicated on the idea that this scaling wouldn’t result in true intelligence. And, as we’ve seen in current reporting from high AI labs like OpenAI and luminaries like François Chollet, the way in which we’ve been approaching deep studying has hit a wall. “Now everyone is looking for the following huge factor,” Sutskever aptly places it. Is it doable that, with strategies like making use of reinforcement studying to LLMs à la OpenAI’s o3, we’re ignoring the knowledge of the bitter lesson (although these strategies are undoubtedly computationally intensive)? What if we sought to know a “principle of all the things” for studying, after which double down on that?
Now we have to deconstruct, then reconstruct, how AI fashions are skilled
Fairly than black-box approximations, at UCSD we developed breakthrough know-how that understands how neural networks truly study. Deep studying fashions characteristic synthetic neurons vaguely much like ours, filtering knowledge by them after which backpropagating them again as much as study options within the knowledge (the latter step is alien to biology). It’s this characteristic studying mechanism that drives the success of AI in fields as disparate as finance and healthcare.
