3 ms·

An AI Which Imitates Humans Can Beat Humans

• haltist 3y ago
  Part of an abstract from referenced articles: > Even though we are able to increase accuracy on MATH, our results show that accuracy remains relatively low, even with enormous Transformer models. Moreover, we find that simply increasing budgets and model parameter counts will be impractical for achieving strong mathematical reasoning if scaling trends continue. While scaling Transformers is automatically solving most other text-based tasks, scaling is not currently solving MATH. To have more traction on mathematical problem solving we will likely need new algorithmic advancements from the broader research community.[1] 1: https://arxiv.org/abs/2103.03874 https://arxiv.org/abs/2103.03874
  • K0balt 3y ago
    I can’t help but smirk at the idea of highly intelligent computers being fundamentally bad at math.
    • haltist 3y ago
      It is very ironic.
• michaelhoney 3y ago
  In addition to the interestingness of the content, this is a nice, readable page hosted on GitHub. Chapeau!