Presentation_1_Detecting Changes and Avoiding Catastrophic Forgetting in Dynamic Partially Observable Environments.pdf (158.94 kB)
Download file

Presentation_1_Detecting Changes and Avoiding Catastrophic Forgetting in Dynamic Partially Observable Environments.pdf

Download (158.94 kB)
presentation
posted on 23.12.2020, 13:28 authored by Jeffery Dick, Pawel Ladosz, Eseoghene Ben-Iwhiwhu, Hideyasu Shimadzu, Peter Kinnell, Praveen K. Pilly, Soheil Kolouri, Andrea Soltoggio

The ability of an agent to detect changes in an environment is key to successful adaptation. This ability involves at least two phases: learning a model of an environment, and detecting that a change is likely to have occurred when this model is no longer accurate. This task is particularly challenging in partially observable environments, such as those modeled with partially observable Markov decision processes (POMDPs). Some predictive learners are able to infer the state from observations and thus perform better with partial observability. Predictive state representations (PSRs) and neural networks are two such tools that can be trained to predict the probabilities of future observations. However, most such existing methods focus primarily on static problems in which only one environment is learned. In this paper, we propose an algorithm that uses statistical tests to estimate the probability of different predictive models to fit the current environment. We exploit the underlying probability distributions of predictive models to provide a fast and explainable method to assess and justify the model's beliefs about the current environment. Crucially, by doing so, the method can label incoming data as fitting different models, and thus can continuously train separate models in different environments. This new method is shown to prevent catastrophic forgetting when new environments, or tasks, are encountered. The method can also be of use when AI-informed decisions require justifications because its beliefs are based on statistical evidence from observations. We empirically demonstrate the benefit of the novel method with simulations in a set of POMDP environments.

History