Autoresearch: Agents researching on single-GPU nanochat training automatically (github.com)

The only thing missing is for the agents to publish and peer-review their research.

Wow, Gemini suggested a very similar experiment to me yesterday. Guess I know where it got the idea from, now. :-)

[dead]

Non-zero based chart makes it look like it was very successful.

[flagged]

but the experiments it did that "improved" validation BPB in the GH screenshot were all basically hyperparameter changes right? So is this better or worse, either per experiment or per unit time, than hyperparameter tuning techniques that don't involve an LLM? It's not clear from this if the LLM is more or less making random changes which sometimes work , and or the LLM thinking actually finds "good" changes because of what the LLM has internalized. E.g. how does this compare to a hyperparameter tuning pass with e.g. BayesOpt that does the same number of 5-min training experiments?

[flagged]

As ai improves, most tasks will become something like this. Environments setup where the model learns through trial and error

Any human endeavor that can be objectively verified in some environment like this can be completely automated

Is there a Autoresearch for Jupyter somewhere? I point it to a Jupyter cell to improve based on another which calculates the target metric?

Would it make this exercise even more interesting if we add that for every 25%+ improvement in val_bpb, existing limits (5 minute and VRAM usage) are also increased (by certain percentages)? This can simuate human-like dev iterations much more closely. Infra can be auto-scaled using a platform like Modal.

> this means that autoresearch will find the most optimal model for your platform in that time budget

I'm looking forward to finding out what model is optimal on my rtx3090

One thing I'm concerned with is that the model with best bpb after 5 minutes in smaller setups are only about ~10M Parameters in size which is too small for some emergent effects.

[dead]

I am in the process of figuring out how to do something similar but to teach a robotic arm a new task in the physical world for ko-br: https://ko-br.com/

I like how it runs out of ideas at the end and just changes the random seed

[dead]

Ah here we go again, the Brophet has unleashed another Brophecy. He seems to confuse brute force discovery with research. Only one leads to understanding, the other one is a shrine to Goodharts law.

How is this different from AlphaEvolve?

https://en.wikipedia.org/wiki/AlphaEvolve

Up next: auto-autoresearch, LLMs searching for autoresearch harnesses and prompts that produce the best results

Goedel machine.

This looks very much like whirlpool. LLM researcher makes LLMs researching LLMs. The quote from old post from Karpathy [1] look very appropriate here

[1] https://karpathy.github.io/2015/05/21/rnn-effectiveness/

  "In particular, setting temperature very near zero will give the most likely thing that Paul Graham might say:
    “is that they were all the same thing that was a startup is that they were all the same thing that was a startup is that they were all the same thing that was a startup is that they were all the same”
  looks like we’ve reached an infinite loop about startups."

As if Karpathy made an artificial Karpathy-researcher-blogger and set temperature close to zero.

Once this can run on stock hardware, set the goal to be replicating to other machines. You get a nice, massively parallel, intelligent guided evolution algorithm for malware. It could even "learn" how to evade detection, how to combine approaches of existing viruses, how to research attack methods, how to identify and exploit vulnerabilities in open source libraries, how to phish, how to blackmail, etc. Maybe even learns how to coordinate attacks with other instances of itself or "publish" new attacks on some encrypted feed it creates. Who knows, maybe it becomes so rampant that instances have to start fighting each other for compute resources. Or maybe eventually one branch becomes symbiotic with humans to fight off their enemies, etc.

[dead]

Adapted this for adversarial protocol hardening. Same loop: markdown defines formal invariants (scope narrowing, cascade revocation), AI tries to violate them, writes tests for whatever breaks. Found compound edge cases that 359 hand-written tests missed, specifically where scope escalation and spend limit bypass interact simultaneously. Property-based testing (100 random inputs per invariant) pairs well with the pattern.

I wonder what happens if I apply the same strategy to an automated shop. Claude code periodically proposes updates and automatically implements them, with revenue as the target function.I'll give it a try.