Hybrid Learning and Reasoning

Seminar an der Universitšt des Saarlandes, Fachrichtung Informatik,  LSF 133348 

Scope:

This seminar (HyLEAR) is concerned with selected hybrid intelligent systems that combine techniques from machine learning with symbolic techniques from knowledge representation and reasoning. Both types of AI techniques have their strengths and limitations. While deep learning systems have been quite successfully applied to, for example, pattern recognition, image interpretation, speech recognition and translation, they can be characterized as overly data hungry, susceptible to adversarial attacks, opaque (non-interpretable by humans), and not informed by general principles such as causality or commonsense and domain knowledge. The successes of symbolic reasoning techniques (“good old fashioned AI”) are in such applications as automated (human-understandable, traceable) planning, diagnosis, design tasks, and question answering by cognitive virtual assistants but are often quite limited by the need of expensive, explicit knowledge acquisition and modelling, inefficient logic-based reasoning, and instability in the presence of noisy data. There is a consensus in the AI community that symbiotic, profound integration or combination of machine learning and reasoning is essential for human-level AI in general. [1,2]

In this seminar, we will take a closer look at selected techniques and systems for hybrid learning and reasoning or planning, and discuss their strengths and weaknesses. The seminar type is classic in the sense that registered participants will present assigned topics, and discuss the strength and weaknesses of presented approaches. In addition, there will be two dedicated opponents for each presentation of an assigned topic. Participation in the seminar will be graded; please check the requirements page in this regard.

The seminar counts 7 ECTS credit points (CS).

Prerequisites:

This seminar aims primarily at advanced master students in Computer Science who preferably hold a B.Sc. degree in this or related field. Good knowledge in AI (introductory course on AI covering symbolic knowledge representation and reasoning, machine or deep learning, automated planning) is required. Selected background references are given on the topic page and expected to be read and utilized by registered participants as appropriate. Attendance of the seminar without registration (no presentation and certificate) by anyone who is interested in the topics is, of course, very much welcome. The seminar language is English or German depending on the audience.

Date and Location:

The seminar is held on Fridays from 2:15 pm to 4 pm.

PLEASE NOTE: First session (introduction with topic assignments) on Monday 25.10.2021 from 4 pm to 6 pm.

Due to the COVID19 pandemic:

  • The seminar takes place virtually in the Web via Zoom unless otherwise stated in the schedule.
  • Each session has an individual online meeting  room address, which will be announced by the seminar organizers on time (see schedule) per E-Mail to all registered participants of the seminar; other interested but not registered parties who like to attend the seminar will be informed on their request.

Please check the seminar schedule frequently for changes.

Registration and Topic Assignments:

Registration request with preferences to the seminar is through the central SIC seminar system until October 19, 2021. Central assignment of students to the seminar will then automatically be done by the system on October 20 22, 2021. If there are free slots available after that date, students can apply for these slots by e-mail to and decision by the seminar organizers.

Assignment of seminar presentation topics will be done during the first session of the seminar on Monday, October 25, 2021, 4 - 6pm. It is recommended to carefully check the reference papers and preliminary presentation dates of topics (see schedule) before registering to the seminar in the central SIC seminar system.

[1] van Harmelen, F. & Teije, A.T. (2019): A Boxology of Design Patterns for Hybrid Learning and Reasoning Systems. Journal of Web Engineering, 18(1-3)
[2] van Bekkum, M., de Boer, M., van Harmelen, F. et al. (2021): Modular design patterns for hybrid learning and reasoning systems: A taxonomy, patterns and use cases. Applied Intelligence.