In this course, we will study human-like artificial agents, that is autonomous intelligent agents situated in a virtual environment similar to real world that act like humans. The course gives an overview of types of such agents and their architectures with the emphasis on the problem of action selection. The course also focuses on solving practical issues related to real-time and partially observable environments. The course is taught at MFF UK as NAIL068.
Lectures: Tuesdays, 10:40, S9 (we start 1.3.2016)
Plan & Slides
- Introduction (Slides 1.1, Slides 1.2 – Gardener Rules)
- Reactive Planning I (Slides 2.1, Slides 2.2)
- Steerings (Slides)
- Environment Awareness (Slides 4.1, Slides 4.2)
- Reactive Planning II (Slides 5.1, Slides 5.2)
- Neural networks for action selection (Slides)
- Ethologically inspired architectures for action selection I (Slides)
- Ethologically inspired architectures for action selection II (Slides in Czech)
- Level-of-detail AI, episodic memory
- Kingdom Come: Deliverance
Experiment (for both MFF & FF UK students)
Attending the experiment is mandatory! You will receive no credits if you miss this.
Experiment takes up to 90-120 minutes. Experiments will start by the end of March and the most of them will happen during April.
For signing-up to the experiment, use orsee laboratory Labels system. First, you have to REGISTER; tick “MFF 2016 LS” during the registration. Whenever you encounter “Experimenty v laboratoři” checkbox, tick it. Then you will start receiving invitations with dates and you will be able to pick one that will work for you.
Experiment report (why we have been doing it and how it connects to the course, etc.) will happen at the end of this semester; you will receive an email with invitation (it will probably happen during one of lectures).
Rules for participating within the experiment:
- I will come fully awake, not after a midnight after-party,
- I will attend only if I do not feel stressed (e.g. by having an exam the very same day, not in a hurry to catch a train home, etc.),
- I do not talk what happened during the experiemtn (do not spoil it for others),
- I sign up to concrete date only if I’m sure to come with at least 90% probability,
- if I get ill, I sing out not to block the date for others (the number of experiment dates is limited),
- you will be required to play RPG game and write some notes; if you cannot write or play, mail the lecturer.
Old Slides (2010-2015)
- Introduction (in Czech)
- Reactive planning, If-then rules, Finite state machines, POSH (updated 130227)
- Pathfinding (in Czech)
- Steering (updated spring 2012)
- Creatures, neural networks, evolutionary algorithms
- Tyrrell (free-flow hierarchy)
- Computational ethology (in Czech)
- Fuzzy approach, emotions (Champandard)
- Belief Desire Intention
- Representation – logic, RETE, affordances, deictic representation
- Agents vs. Animats, Wooldridge, FIPA, speech acts
- Soar intro
- Storytelling intro
- Spatial memory & psychological experiments (updated 130425)
- Slides on emotions (2007, in Czech)