Thursday, January 20, 2022

Prospective Learning: Back to the Future

Abstract for the linked article:

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.


  1. " FBI investigators did find that cells of protesters, including followers of the far-right Oath Keepers and Proud Boys groups, had aimed to break into the Capitol. But they found no evidence that the groups had serious plans about what to do if they made it inside"

    F.B.I have found no evidence of sedition argument! Based on an old arrest pattern suspects accused all needed to be in the space entering the building.

    Failing to take on board that with this conspiracy where you are is irrelevant, that is the game changer here.

    Plus not noting that being in the building and planning to obstruct the transfer of power is central but only one component of the charge.

    Its obviously 'difficult' to persuade pro Trump supporters but when people who are very much opposed to Trump cant see and accept the premise, highly effective disinformation.

  2. 'Western scrub-jays not only have a complex spatial memory of food cache locations, but can flexibly adapt their caching to anticipate future needs.'

    Got the memory but not getting the data to shift and alter unless you actively parse the data for yourself and do the work.

    Back to the future issue very much a part of Americas political difficulty and disinformation issue.