Skip to main content

Indie game storeFree gamesFun gamesHorror games
Game developmentAssetsComics
SalesBundles
Jobs
TagsGame Engines

I really resonate with the core of your pitch here. I also agree that the field is stuck, biology works, and the gap between the two is underexplored. And the fact that you've actually built something that demonstrates sparse reward learning with a hippocampal memory system puts this well above a pure ideas pitch. The Snake results are genuinely interesting.

A few questions I was left with after reading your submission:

You name regional specialization as what differentiates your approach from the cargo cult biomimetic efforts you critique. But regional specialization is itself a mechanical specific of the brain, so what's the principled story for why this is the key insight those other efforts missed, rather than just being a different flavor of bio-mimicry? 

Your stated vision of success is frontier labs redirecting resources toward alternative architectures. But the method is "I'll build a small proof of concept that shocks them into action." How would you evaluate whether what you're doing is actually moving you toward that institutional goal? And is building it yourself necessarily the most efficient path there, versus say, writing the theoretical case, getting embedded at a lab, or amplifying existing aligned work like Sakana's CTM? Developing your own novel biomimetic architecture that's impressive enough to get frontier labs to start allocating their resources away from their current research efforts and into more novel biomimetic approaches is something that'll take a lot of time and resources to invest in, but I don't see how you're keeping a metric of evaluating if you're moving closer to success or not. 

Lastly, what's the failure model? The pitch reads as a straight line from Snake to Atari 100k to drone racing, but bio-inspired work is full of unexpected walls. If Atari 100k doesn't go well, how do you diagnose whether the problem is in your implementation or your thesis or your publicity strategy? What does a pivot look like? If Atari 100k does go well and it doesn't receive the attention you expected, how do you move from there? 

Overall, the pitch is pretty solid. The conviction is clear, the direction is worth exploring, and you're already developing something novel and interesting. Just want to see a sharper account of why your specific approach is the right one, and a plan for what happens if the road gets bumpy.

Thank you!

My intuition, and what the neuroscience suggests, is that specialized brain regions are each providing specific services to the overall network which allow the sum of behavior to be flexible, persistent, and efficient to train. It seems to me that we have so far done a pretty good job of building ML networks that do a good job of doing the individual tasks of specific subsets of the overall brain -- CNNs are great at visual perception, we have fantastic speech transcription models, and we're increasingly good at language processing, obviously. The thing I think is missing is a network approach where we try to understand what each region of the brain is doing, how that service integrates to the broader picture, and how those interactions can be meaningfully captured in code. It's worth pointing out that the proof of concept can be made at arbitrarily small scales. Some species of parasitoid wasps have fully functional brains, capable of navigating them in flight, with just a few thousand neurons. And, of course, C. Elegans, everyone's favorite model worm, with its ~300 neurons, is a perfectly functional organism. It should be possible to prove cohesive integration of all brain regions into a useful and persistent entity, at a very small scale. But maybe I'm wrong, and the answer really is "Just keep scaling deep, amorphous networks."

You're right that I am really just taking a bet on the direction that I think has the highest chance of making an impact. The reasoning behind my belief is that there is essentially an infinite amount of value locked behind robotic AI, and obviously all the labs want it. We have a ton of companies working on building robots, and they're all trying to run them but struggling. If I can build a network which does even marginally better at practical robotics than the other architectures, I think that would get attention. If it does substantially better at robotics, especially running on edge hardware, it seems to me that it would be guaranteed to get a *lot* of attention. Again, we have the robots, and all the decent people want the infinite production, post-scarcity future ASAP. That future is quite literally held up on practical robotics, and I'm hoping to take steps in that direction, in a way that helps the whole industry head that way.

Yep, the failure mode is fundamentally that I'm not able to make any gains compared to existing architectures, in video game playing, drone navigation, or general environment interaction. If I try my best for a good while, and I just can't figure out how to do any of the things I'm envisioning, I will have failed, and I'll have to move on. Though, for the last case, there's actually a very different story - if Atari 100k does actually go quite well, and nobody cares, then I just turn evil and unleash an army of AGIs on the world. Look out, Will Stancil. In seriousness, though, if the architecture does well at video games, I'm fully confident it will scale to robotics, and I'll start buying robots, uploading useful brains into them, and selling them for 20x. Gardening robot? Cooking/washing dishes robot? Cleaning robot? People would pay crazy money. 

Thanks for the feedback!