5 ms·

Very fascinating result, and it seems they wrote this blog post out of pure excitement to share their findings, and maybe to have someone throw cold water on it

Very fascinating result, and it seems they wrote this blog post out of pure excitement to share their findings, and maybe to have someone throw cold water on it before publishing, ha. Who knows if this is the actual fabled path of "self improvement", but results like this are what we expect to find on such a path.

suddenlybananas 1y ago
> Who knows if this is the actual fabled path of "self improvement" Seems doubtful as this works only on an extremely well-defined evaluation function.
- EMIRELADERO 1y ago
  That may be true, but this is the first example I've seen where the concept is successfully implemented in a noticeable way. It's just like image generation: the first iteration is the worst it will ever be.
- observationist 1y ago
  Each time you define another task well enough for the system to work, you generalize the system just a little bit - repeat enough times and you can start to expand, develop taxonomies of functions, precisely define function spaces and metrics for improvement. This might not be a bootstrap for recursive self improvement generally, but it could definitely inform the theory or design of a system that does bootstrap rsi.
  - suddenlybananas 1y ago
    That's an entirely different idea that may or may not work. This is not evidence of that.
    - observationist 1y ago
      The structure of their research - the process, the specific task, and the data they generate - will help inform how other research gets performed. Instead of GPU kernels, maybe the next task is something like neuron modules, looking for structures that improve on attention blocks, or things like that - each time you run through an experiment like this, you're creating foundational data upon which other experiments can be run and improved. Once you've done enough of them, you can generalize. It could be that the end result is the knowledge of strict boundaries of LLM capabilities, that they can only operate in specific domains, or only improve to a certain extent, and some currently unspecified defect limits the level of improvement. The underlying idea of specifying a domain and task conditions, then letting an LLM run thousands of experiments, is a great search technique. The hope is that there is no implicit defect and that the methodology will extend and generalize - it's not too complex a notion to think that you could have an LLM create a broad range of individual tasks, with a meta-goal of identifying better and more general recursive improvement processes and algorithms.
      - suddenlybananas 1y ago
        >The hope is that there is no implicit defect and that the methodology will extend and generalize - it's not too complex a notion to think that you could have an LLM create a broad range of individual tasks, with a meta-goal of identifying better and more general recursive improvement processes and algorithms Again, entirely different idea that doesn't have a straightforward evaluation function. As it stands, this is more akin to genetic programming with a very good mutation function.

suddenlybananas 1y ago

> Who knows if this is the actual fabled path of "self improvement" Seems doubtful as this works only on an extremely well-defined evaluation function.

EMIRELADERO 1y ago
That may be true, but this is the first example I've seen where the concept is successfully implemented in a noticeable way. It's just like image generation: the first iteration is the worst it will ever be.
observationist 1y ago
Each time you define another task well enough for the system to work, you generalize the system just a little bit - repeat enough times and you can start to expand, develop taxonomies of functions, precisely define function spaces and metrics for improvement. This might not be a bootstrap for recursive self improvement generally, but it could definitely inform the theory or design of a system that does bootstrap rsi.
- suddenlybananas 1y ago
  That's an entirely different idea that may or may not work. This is not evidence of that.
  - observationist 1y ago
    The structure of their research - the process, the specific task, and the data they generate - will help inform how other research gets performed. Instead of GPU kernels, maybe the next task is something like neuron modules, looking for structures that improve on attention blocks, or things like that - each time you run through an experiment like this, you're creating foundational data upon which other experiments can be run and improved. Once you've done enough of them, you can generalize. It could be that the end result is the knowledge of strict boundaries of LLM capabilities, that they can only operate in specific domains, or only improve to a certain extent, and some currently unspecified defect limits the level of improvement. The underlying idea of specifying a domain and task conditions, then letting an LLM run thousands of experiments, is a great search technique. The hope is that there is no implicit defect and that the methodology will extend and generalize - it's not too complex a notion to think that you could have an LLM create a broad range of individual tasks, with a meta-goal of identifying better and more general recursive improvement processes and algorithms.
    - suddenlybananas 1y ago
      >The hope is that there is no implicit defect and that the methodology will extend and generalize - it's not too complex a notion to think that you could have an LLM create a broad range of individual tasks, with a meta-goal of identifying better and more general recursive improvement processes and algorithms Again, entirely different idea that doesn't have a straightforward evaluation function. As it stands, this is more akin to genetic programming with a very good mutation function.