Effects of Input Addition in Learning for Adaptive Games: Towards Learning with Structural Changes

Iago Bonnici 1 Abdelkader Gouaich 1 Fabien Michel 1
1 SMILE - Système Multi-agent, Interaction, Langage, Evolution
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier
Abstract : Adaptive Games (AG) involve a controller agent that continuously feeds from player actions and game state to tweak a set of game parameters in order to maintain or achieve an objective function such as the flow measure defined by Csíkszentmihályi. This can be considered a Reinforcement Learning (RL) situation, so that classical Machine Learning (ML) approaches can be used. On the other hand, many games naturally exhibit an incremental gameplay where new actions and elements are introduced or removed progressively to enhance player's learning curve or to introduce variety within the game. This makes the RL situation unusual because the controller agent input/output signature can change over the course of learning. In this paper, we get interested in this unusual "protean" learning situation (PL). In particular, we assess how the learner can rely on its past shapes and experience to keep improving among signature changes without needing to restart the learning from scratch on each change. We first develop a rigorous formalization of the PL problem. Then, we address the first elementary signature change: "input addition", with Recurrent Neural Networks (RNNs) in an idealized PL situation. As a first result, we find that it is possible to benefit from prior learning in RNNs even if the past controller agent signature has less inputs. The use of PL in AG thus remains encouraged. Investigating output addition, input/output removal and translating these results to generic PL will be part of future works.
Complete list of metadatas

Cited literature [26 references]  Display  Hide  Download

https://hal-lirmm.ccsd.cnrs.fr/lirmm-02127618
Contributor : Iago Bonnici <>
Submitted on : Monday, May 13, 2019 - 3:50:47 PM
Last modification on : Wednesday, May 15, 2019 - 5:43:45 AM

File

submitted_version.pdf
Files produced by the author(s)

Identifiers

Collections

Citation

Iago Bonnici, Abdelkader Gouaich, Fabien Michel. Effects of Input Addition in Learning for Adaptive Games: Towards Learning with Structural Changes. EvoApplications: Applications of Evolutionary Computation, Apr 2019, Leipzig, Germany. pp.172-184, ⟨10.1007/978-3-030-16692-2_12⟩. ⟨lirmm-02127618⟩

Share

Metrics

Record views

69

Files downloads

65