Jordi Bieger, MSc
Jordi Bieger, MSc is currently working as a Researcher at Delft University of Technology and is a Ph.D. candidate at Reykjavik University where his research focus is on Artificial Intelligence. At Delft, he is implementing the Massive Open Online Deliberation (MOOD) platform that aims to enhance critical thinking and reflection among discussion participants by providing a formalized and guided process of moral deliberation.
Additionally, Jordi is conducting research on Artificial (General) Intelligence and Artificial Pedagogy, the study of teaching, training, raising, and educating AI systems, with a particular eye towards Artificial General Intelligence.
He wants to develop a theory for how different teaching techniques such as demonstration, heuristic rewarding, simplification, and decomposition can help AI systems to learn. How can we shape the environment and present training data? How can we interactively determine learning progress and remedy deficiencies? To answer these questions we need a Task Theory for modeling tasks, and AI Evaluation for (interactively) determining the growing capabilities of learning systems.
At Reykjavik University, Jordi was also a Teaching Assistant on Advanced Topics in AI (2016–2017) and on Research Methodology (2014–2015).
Jordi’s related research interests include Transfer Learning, Lifelong Learning, Multitask Learning, Multiagent Learning, Semi-supervised Learning, Active Learning, Computational Learning Theory, and Reinforcement Learning.
He is mainly interested in advanced, general, safe, and beneficial artificial intelligence for all. Currently he is working on how AI can be used ethically for law enforcement, and is helping to implement the new MOOD platform for online value-driven deliberation. He also aided in developing and running the Mind of the Universe — Robots in Society: Blessing or Curse? online learning experience on EdX.
In 2015, Jordi, together with Ben Goertzel and Alexey Potapov, Coedited the book Artificial General Intelligence: 8th International Conference, AGI 2015, Berlin, Germany, July 22–25, 2015, Proceedings (Lecture Notes in Computer Science). The AGI conference series has played and continues to play, a significant role in the resurgence of research on artificial intelligence in the deeper, original sense of the term artificial intelligence. Jordi was also the Program Committee Co-Chair for AGI-15, the 8th Conference on Artificial General Intelligence.
Jordi was the Coorganizer of the AEGAP 2018 Workshop on Architectures and Evaluation for Generality, Autonomy & Progress in AI at the Federated AI Meeting, in Stockholm and Coorganizer of the AGA 2017 Workshop on Architectures for Generality & Autonomy at IJCAI-17 in Melbourne.
In 2016, he coauthored a research paper Why Artificial Intelligence Needs a Task Theory — And What It Might Look Like, which is part of a series on the GoodAI webpage whose mission is to develop general artificial intelligence — as fast as possible — to help humanity and understand the universe.
Before Jordi began the pursuit of his Ph.D. in 2013, he was Research Assistant at the Icelandic Institute for Intelligent Machines, working on reinforcement learning and its evaluation in multi-objective and continuous domains.
Between 2010 and 2013, he was Vision Engineer at VicarVision, where they were developing state of the art solutions and products for computer vision.
Jordi earned his Master’s Degree of Science in Artificial Intelligence in 2010 from Radboud University Nijmegen. In 2005, after a year of attendance, he became a Teaching Assistant, helping to create the practical assignments and assist students, and occasionally take over the lectures. While studying in 2007, he also worked as a Programmer at RE-phrase creating web applications and in the last year of studies, he worked as a Student Researcher at Philips in Eindhoven. Jordi worked on his graduation internship in the Brain, Body & Behavior group at Philips Research. He is focused on state-of-the-art brain computer interfaces (BCIs).
In his free time, Jorid also practices Taekwon-Do, and for two years, he was also a Taekwond-Do Teacher at Taekwon-Do Vereniging Chang-Hun.
Read Jordi’s AGI 2014 Conference presentation and paper of his work on Raising AI: Tutoring Matters or watch it on YouTube.
Watch his video presentation on Universal Pedagogy and Teaching AI Systems.
Watch the Panel discussion including Jordi and his presentation about Task Analysis for Teaching Cumulative Learners at Human-Level AI 2018.
Visit his LinkedIn profile, TUDelft profile, his homepage, Google Scholar page, ResearchGate profile, and dblp page. Follow him on Facebook, Semantic Scholar, Academia, and Twitter.
