5 ms·

Using approaches like OpenAI's recent evolution strategies paper would remove the need for backprop, likely allowing these TPUs to be using for training without

by sgk284 10y ago

Using approaches like OpenAI's recent evolution strategies paper would remove the need for backprop, likely allowing these TPUs to be using for training without any changes.

p1esk 10y ago
Evolution strategy method is used in reinforcement learning models. How are you planning to use it for supervised learning?
- mdda 10y ago
  People have known that training NNs (for any purpose) using evolution works well since the 1990s. The rise of the NN frameworks has made doing differentiation much easier now than it was before (and having gradient hints is intuitively a good idea). But for OpenAI to allow their PR people to declare this as a novel advance is ... surprising.
  - p1esk 10y ago
    Citation for training NN for image classification task where evolution works well? Let's say you want to use a genetic algorithm to find a good set of weights: you generate, mutate, combine and select many random networks, and repeat this process many times. How many networks and how many times? That depends on the length of your chromosome and complexity of the task. Networks that work well for image classification need at least a million weights. The entire set of weights is a single chromosome. You realize now how computationally intractable this task is on modern hardware?
    - mdda 10y ago
      > NN for image classification task You've created your own straw man here. > "You realize now how computationally intractable this task is on modern hardware?" Here are the people that prove it isn't computationally intractable : https://blog.openai.com/evolution-strategies/ https://blog.openai.com/evolution-strategies/ - but to say they've discovered a new breakthrough method is over-selling the result.
      - p1esk 10y ago
        You said: "training NNs (for any purpose) using evolution works well". I gave you an example of a purpose where it does not work well. So, let me ask you again: can you give an example of evolutionary methods that work well when applied to training NNs, other than this breakthrough by OpenAI, which only works for RL?

by sgk284 10y ago

Using approaches like OpenAI's recent evolution strategies paper would remove the need for backprop, likely allowing these TPUs to be using for training without any changes.

p1esk 10y ago

Evolution strategy method is used in reinforcement learning models. How are you planning to use it for supervised learning?

mdda 10y ago
People have known that training NNs (for any purpose) using evolution works well since the 1990s. The rise of the NN frameworks has made doing differentiation much easier now than it was before (and having gradient hints is intuitively a good idea). But for OpenAI to allow their PR people to declare this as a novel advance is ... surprising.
- p1esk 10y ago
  Citation for training NN for image classification task where evolution works well? Let's say you want to use a genetic algorithm to find a good set of weights: you generate, mutate, combine and select many random networks, and repeat this process many times. How many networks and how many times? That depends on the length of your chromosome and complexity of the task. Networks that work well for image classification need at least a million weights. The entire set of weights is a single chromosome. You realize now how computationally intractable this task is on modern hardware?
  - mdda 10y ago
    > NN for image classification task You've created your own straw man here. > "You realize now how computationally intractable this task is on modern hardware?" Here are the people that prove it isn't computationally intractable : https://blog.openai.com/evolution-strategies/ https://blog.openai.com/evolution-strategies/ - but to say they've discovered a new breakthrough method is over-selling the result.
    - p1esk 10y ago
      You said: "training NNs (for any purpose) using evolution works well". I gave you an example of a purpose where it does not work well. So, let me ask you again: can you give an example of evolutionary methods that work well when applied to training NNs, other than this breakthrough by OpenAI, which only works for RL?