Abstract
An important aspect of intelligence is the ability to adapt to a novel task without any direct experience (zero shot), based on its relationship to previous tasks. Humans can exhibit this cognitive flexibility. By contrast, models that achieve superhuman performance in specific tasks often fail to adapt to even slight task alterations. To address this, we propose a general computational framework for adapting to novel tasks based on their relationship to prior tasks. We begin by learning vector representations of tasks. To adapt to new tasks, we propose metamappings, higher-order tasks that transform basic task representations. We demonstrate the effectiveness of this framework across a wide variety of tasks and computational paradigms, ranging from regression to image classification and reinforcement learning. We compare to both human adaptability and language-based approaches to zero-shot learning. Across these domains, metamapping is successful, often achieving 80 to 90% performance, without any data, on a novel task, even when the new task directly contradicts prior experience. We further show that metamapping can not only generalize to new tasks via learned relationships, but can also generalize using novel relationships unseen during training. Finally, using metamapping as a starting point can dramatically accelerate later learning on a new task and reduce learning time and cumulative error substantially. Our results provide insight into a possible computational basis of intelligent adaptability and offer a possible framework for modeling cognitive flexibility and building more flexible artificial intelligence systems.
Funder
National Science Foundation
Publisher
Proceedings of the National Academy of Sciences
Reference55 articles.
1. Turing on super-Turing and adaptivity;Siegelmann;Prog. Biophys. Mol. Biol.,2013
2. Building machines that learn and think like people
3. G. Marcus , Deep learning: A critical appraisal. arXiv:1801.00631 (2 January 2018).
4. J. Russin , R. C. O’Reilly , Y. Bengio , “Deep learning needs a pre-frontal cortex” in ICLR Workshop on Bridging AI and Cognitive Science. https://baicsworkshop.github.io/. Accessed 26 April 2020.
5. On the control of automatic processes: A parallel distributed processing account of the Stroop effect.
Cited by
6 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献