Jul 10

Measuring abstract reasoning in neural networks

Whether neural networks can learn abstract reasoning or whether they merely rely on superficial statistics is a topic of recent debate. Here, we propose a dataset and challenge designed to probe abstract reasoning, inspired by a well-known human IQ test. To succeed at this challenge, models must cope with various generalisation ‘regimes’ in which the training and test data differ in clearly-defined ways. We show that popular models such as ResNets perform poorly, even when the training and test sets differ only minimally, and we present a novel architecture, with a structure designed to encourage reasoning, that does significantly better. When we vary the way in which the test questions and training data differ, we find that our model is notably proficient at certain forms of generalisation, but notably weak at others. We further show that the model’s ability to generalise improves markedly if it is trained to predict symbolic explanations for its answers. Altogether, we introduce and explore ways to both measure and induce stronger abstract reasoning in neural networks. Our freely-available dataset should motivate further progress in this direction.

Read full article >

Dec 22

MuZero: Mastering Go, chess, shogi and Atari without rules

In a paper in the journal Nature, we describe MuZero, a significant step forward in the pursuit of general-purpose algorithms. MuZero masters Go, chess, shogi and Atari without needing to be told the rules, thanks to its ability to plan winning strategies in unknown environments.

Read full article >

Dec 03

Using JAX to accelerate our research

We've found that an increasing number of projects are well served by JAX, a machine learning framework developed by Google Research teams. JAX resonates well with our engineering philosophy and has been widely adopted by our research community over the last year.

Read full article >

Nov 29

AlphaFold: a solution to a 50-year-old grand challenge in biology

In a major scientific advance, the latest version of our AI system AlphaFold has been recognised as a solution to the protein folding problem by the organisers of the biennial Critical Assessment of protein Structure Prediction (CASP).

Read full article >