• Posts
Artificial Intelligence
• Reinforcement Learning
• Automated Planning
  • BSc Thesis I
  • BSc Thesis II
  • BSc Thesis III
  • BSc Thesis IV

BSc Thesis IV

Domain-Dependent Policy Learning using Neural Networks for Classical Planning (4/4) This will be the forth and final post of my undergraduate dissertation series. It will cover the detailed evaluation of Action Schema Networks conducted for classical automated planning, propose future work that might deal with identified weaknesses before concluding the project as a whole. Evaluation As mentioned in the second post of this series, Sam Toyer already conducted an empirical evaluation of ASNets [1], but primarily focused on probabilistic planning tasks.

June 26, 2019 Read

BSc Thesis III

Domain-Dependent Policy Learning using Neural Networks for Classical Planning (3/4) This third post about my undergraduate dissertation will cover my primary contributions to translate the architecture of Action Schema Networks, introduced in the previous post, for classical automated planning in the Fast-Downward framework. The dissertation focuses on the application of ASNets in deterministic, classical planning. For this purpose, the network architecture was implemented and integrated into the Fast-Downward planning system [1] which is prominently used throughout classical planning research.

June 2, 2019 Read

BSc Thesis II

Domain-Dependent Policy Learning using Neural Networks for Classical Planning (2/4) This post as the second of the series about my undergraduate thesis will cover the underlying architecture of Action Schema Networks. Action Schema Networks Action Schema Networks, short ASNets, is a neural networks architecture proposed by Sam Toyer et al. [1, 2] for application in automated planning. The networks are capable of learning domain-specific policies to exploit on arbitrary problems of a given (P)PDDL domain.

May 5, 2019 Read

BSc Thesis I

Domain-Dependent Policy Learning using Neural Networks for Classical Planning (1/4) I have finished my undergraduate Bachelor studies last summer and as a start to this blog I will outline the work I did for my dissertation titled Domain-Dependent Policy Learning using Neural Networks in Classical Planning I will split this summary over four posts which will mostly be constructed of paragraphs of my thesis, summaries of such or parts of the kolloquium presentation I held at the group seminar of the Foundations of Artificial Intelligence (FAI) groupat Saarland University.

December 3, 2018 Read