AlphaTensor: discovering mathematical algorithms with reinforcement learning

Go back to programme

AlphaTensor: discovering mathematical algorithms with reinforcement learning

  • Watch

    * Register (or log in) to the AI4G Neural Network to add this session to your agenda or watch the replay

    • Watch

      Register

      * Register (or log in) to the AI4G Neural Network to add this session to your agenda or watch the replay

    Improving the efficiency of algorithms for fundamental computations has a widespread impact, as it affects the overall speed of a very large amount of computations. Matrix multiplication is one such primitive task, occurring in many systems—from neural networks to scientific computing routines. The automatic discovery of algorithms using machine learning offers the prospect of reaching beyond human intuition and outperforming the current best human-designed algorithms. However, automating the algorithm discovery procedure is intricate, as the space of possible algorithms is enormous.  

    This AI for Good session presents a deep reinforcement learning approach based on AlphaZero for discovering efficient and provably correct algorithms for the multiplication of arbitrary matrices. The agent, AlphaTensor, is trained to play a single-player game where the objective is finding tensor decompositions within a finite factor space. AlphaTensor discovered algorithms that outperform the state-of-the-art complexity for many matrix sizes. Particularly relevant is the case of 4×4 matrices in a finite field, where AlphaTensor’s algorithm improves on Strassen’s two-level algorithm for the first time, to our knowledge, since its discovery 50 years ago.  

    This talk will further showcase the flexibility of AlphaTensor through different use-cases: algorithms with state-of-the-art complexity for structured matrix multiplication and improved practical efficiency by optimizing matrix multiplication for runtime on specific hardware. Our results highlight AlphaTensor’s ability to accelerate the process of algorithmic discovery on a range of problems, and to optimize for different criteria.

    This live event includes a 15-minute networking event hosted on the AI for Good Neural Network. This is your opportunity to ask questions, interact with the panelists and participants and build connections with the AI for Good community.

    Share this session
    • Watch

      Register

      * Register (or log in) to the AI4G Neural Network to add this session to your agenda or watch the replay

    • Start date
      12 January 2023 at 17:00 CET Geneva | 11:00-12:15 EST, New York | 23:00-00:15 CST, Beijing
    • End date
      12 January 2023 at 18:15 CET Geneva | 11:00-12:15 EST, New York | 23:00-00:15 CST, Beijing
    • Duration
      75 minutes (including 15 minutes networking)
    • Topics
    • UN SDGs

    Are you sure you want to remove this speaker?