Prospective Learning in Retrospect
This work addresses the challenge of adapting machine learning to real-world scenarios where data and goals change over time, but it appears incremental as it builds on an existing framework.
The paper tackles the problem of dynamic data distributions and evolving objectives in AI, which the PAC learning framework fails to address, by building on the prospective learning framework to improve algorithms and extend it to sequential decision-making like foraging, showing preliminary improvements in performance.
In most real-world applications of artificial intelligence, the distributions of the data and the goals of the learners tend to change over time. The Probably Approximately Correct (PAC) learning framework, which underpins most machine learning algorithms, fails to account for dynamic data distributions and evolving objectives, often resulting in suboptimal performance. Prospective learning is a recently introduced mathematical framework that overcomes some of these limitations. We build on this framework to present preliminary results that improve the algorithm and numerical results, and extend prospective learning to sequential decision-making scenarios, specifically foraging. Code is available at: https://github.com/neurodata/prolearn2.