The CLRS Algorithmic Reasoning Benchmark
This work addresses the problem of fragmented evaluation in algorithmic reasoning for researchers, providing a consolidated benchmark to facilitate progress, though it is incremental as it builds on existing methods by standardizing data.
The authors tackled the lack of unified evaluation in algorithmic reasoning by proposing the CLRS Algorithmic Reasoning Benchmark, which covers classical algorithms from a textbook and includes extensive experiments to benchmark popular baselines and highlight open challenges.
Learning representations of algorithms is an emerging area of machine learning, seeking to bridge concepts from neural networks with classical algorithms. Several important works have investigated whether neural networks can effectively reason like algorithms, typically by learning to execute them. The common trend in the area, however, is to generate targeted kinds of algorithmic data to evaluate specific hypotheses, making results hard to transfer across publications, and increasing the barrier of entry. To consolidate progress and work towards unified evaluation, we propose the CLRS Algorithmic Reasoning Benchmark, covering classical algorithms from the Introduction to Algorithms textbook. Our benchmark spans a variety of algorithmic reasoning procedures, including sorting, searching, dynamic programming, graph algorithms, string algorithms and geometric algorithms. We perform extensive experiments to demonstrate how several popular algorithmic reasoning baselines perform on these tasks, and consequently, highlight links to several open challenges. Our library is readily available at https://github.com/deepmind/clrs.