Reasoning Agents (3 CFU)
General Information
One of the most important goals of Artificial Intelligence
concerns the development of intelligent agents, such as robots,
that are able to deliberate their course of actions.
This course introduces several approaches to
realize this goal, with a particular focus on advanced
forms of automated behaviour synthesis, based on temporal logics and
action theories, in both deterministic and non-deterministic
settings.
Students are encouraged to realize projects which integrate
such high-level forms of reasoning, with state-of-the-art
probabilistic robotics, computer vision,
and machine learning tools.
Lectures
Course taught from April 21 to May 27, 2022.
Lectures will take place according to the following schedule:
- Thu 13.00-17.00, room A2 (DIAG)
- Fri 15.00-17.00, room A2 (DIAG)
Lectures will be given in presence and can be attended online
here.
All information about access to Sapienza buldings can be found
here.
Google Classroom
Teacher-student communication will take place on
Google Classroom
(registration code: wxva7ms).
Student hour
Information available
here.
Program
-
Modeling world dynamics: Transition Systems.
-
Checking dynamic properties.
CTL Model Checking, Fixpoints.
-
Planning in deterministic and nondeterministic domains.
Planning for reachability goals in deterministic domains.
Planning for reachability goals in nondeterministic domains.
-
Synthesis for LTLf and LDLf goals.
Linear-time Temporal Logic (LTL).
LTLf and LDLf on finite traces, LTLf and LDLf goals.
DFAs for LTLf/LDLf formulas.
Planning for LTLf and LDLf goals.
News
- April 14 2022: course starts on April 21
Teaching Material
Slides
| Week |
Thursday, 13.00-17.00 |
Friday, 15.00-17.00 |
| 18/04-24/04/2022 |
Introduction. Transition Systems.
Paper presentations.
|
Transition Systems.
|
| 25/04-01/05/2022 |
Paper presentations.
|
Transition Systems.
Computation Tree Logic (CTL) - introduction.
|
| 02-08/05/2022 |
Computation Tree Logic (CTL).
CTL Model Checking.
|
CTL Model Checking.
Finite-state automata (introduction).
|
| 09-15/05/2022 |
Finite-state automata.
|
Linear-time temporal logics on finite traces (LTLf/LDLf).
Planning for LTLf/LDLf.
|
| 16-22/05/2022 |
Seminar (Dr. Alessandro Trapasso):
The AIPlan4EU Unified Planning Framework
Paper presentations.
|
Paper presentations.
|
| 23-29/05/2022 |
Seminar (Dr. Francesco Fuggitti):
Planning for LTLf/LDLf goals.
Paper presentations.
|
Paper presentations.
End of course.
|
Exams
Exam format
The exam consists in two parts:
-
Presentation of a paper related to the topics of the course.
The list of papers will be announced during the course.
Every student must present one paper, different from those of
the other students.
-
Implementation of a project, jointly with HRI, where the
techniques studied in the two courses are applied.
Please check Classroom for further details.
Exam dates
Please check Infostud and Classroom for details.