Prospective Learning: Back to the Future

Abstract

Research on both natural intelligence (NI) and artificial intelligence (AI) generally assumes that the future resembles the past: intelligent agents or systems (what we call 'intelligence') observe and act on the world, then use this experience to act on future experiences of the same kind. We call this 'retrospective learning'. For example, an intelligence may see a set of pictures of objects, along with their names, and learn to name them. A retrospective learning intelligence would merely be able to name more pictures of the same objects. We argue that this is not what true intelligence is about. In many real world problems, both NIs and AIs will have to learn for an uncertain future. Both must update their internal models to be useful for future tasks, such as naming fundamentally new objects and using these objects effectively in a new context or to achieve previously unencountered goals. This ability to learn for the future we call 'prospective learning'. We articulate four relevant factors that jointly define prospective learning. Continual learning enables intelligences to remember those aspects of the past which it believes will be most useful in the future. Prospective constraints (including biases and priors) facilitate the intelligence finding general solutions that will be applicable to future problems. Curiosity motivates taking actions that inform future decision making, including in previously unmet situations. Causal estimation enables learning the structure of relations that guide choosing actions for specific outcomes, even when the specific action-outcome contingencies have never been observed before. We argue that a paradigm shift from retrospective to prospective learning will enable the communities that study intelligence to unite and overcome existing bottlenecks to more effectively explain, augment, and engineer intelligences."No man ever steps in the same river twice. For it's not the same river and he's not the same man." -Heraclitus

Habituation Reflects Optimal Exploration Over Noisy Perceptual Samples

“…This characteristic could constrain their application to situations in which the environment is stable and the cost of sampling is low. Since adult exploration is sensitive to environmental complexity, a forward-looking metric like EIG might be particularly suitable to predict behaviors in a more dynamic learning context (Dubey & Griffiths, 2020;Vogelstein et al, 2022).…”

On the link between conscious function and general intelligence in humans and machines

“…In cases where this context is imagined, the agent would be able to simulate possible policies, each which respond according to the imagined situation, rather than the current one. This ability to take future goals into account is in line with the "prospective" learning seen in natural agents, as opposed to the largely "retrospective" learning implemented in current AI systems (Vogelstein et al, 2022).…”

Naive Few-Shot Learning: Sequence Consistency Evaluation

“…Our focus here is the extent to which naive models can solve this sub-problem, extracting simple rules from sequences without prior training. This is an unusual setting for deep learning models, in which pretraining is considered crucial even in the context of few-shot learning (Chollet, 2019;Vogelstein et al, 2022).…”