DeepMind AlphaCode
Using transformer-based language models to generate code
Über DeepMind AlphaCode
AlphaCode is the first AI code generation system that has “reached a competitive level of performance in programming competitions.”
It uses a large language model to build code in response to natural language descriptions of a problem. The software takes advantage of a massive data set of programming problems and solutions, plus a set of unstructured code from GitHub. AlphaCode generates thousands of proposed solutions to the problem at hand, filters those solutions to toss out the ones that aren’t valid, clusters the solutions that survive into groups, and then selects a single example from each group to submit.
AlphaCode claims it performs better than OpenAI’s GPT-3-powered OpenAI Codex. AlphaCode scored a significant win at a Codeforces programming contest, ranking on average in the top 54% against more than 5,000 participants in 10 contests.
Source: https://www.deepmind.com/blog/competitive-programming-with-alphacode