Transforming task representations to perform novel tasks

By Andrew Kyle Lampinen, James L. McClelland. 2020

18 citations recorded by the source corpus.

Significance An intelligent system should be able to adapt to a novel task without any data and achieve at least moderate success. Humans can often do so, while models often require immense datasets to reach human-level performance. We propose a general computational framework by which models can adapt to new tasks based only on their relationship to old tasks. Our approach is based on transforming learned task representations. Our approach allows models to perform well on novel tasks, even using novel relationships not encountered during training. This adaptation can substantially accelerate later learning. Our work could contribute to understanding the computational basis of intelligence, to cognitive modeling, and to building more flexible forms of artificial intelligence. 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.

Subfields: Artificial Intelligence, Representation & Self-Supervised Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Representation & Self-Supervised Learning

Citations in the corpus

The Power of Scale for Parameter-Efficient Prompt Tuning (2021, background): “to English by manipulating the task prompt for an English to German task with a learned operator that flips the order of the arguments? Learning these operators is active area of research (Lampinen and McClelland, 2020) that would benefit from”
Towards mental time travel: a hierarchical memory for reinforcement learning agents (2021, background): “Recent work shows that transforming prior task representations allows zero-shot adaptation to substantially altered tasks [30].”
Editing Models with Task Arithmetic (2022, unknown): “Additionally, Lampinen & McClelland [50] propose a framework for adapting models based on relationships between tasks.”