Boi Faltings | EPFL Graph Search

Are you an EPFL student looking for a semester project?

Work with us on data science and visualisation projects, and deploy your project as an app on top of GraphSearch.

Professor Faltings joined EPFL in 1987 as professor of Artificial Intelligence. He holds a PhD degree from the University of Illinois at Urbana-Champaign, and a diploma from the ETHZ. His research has spanned different areas of intelligent systems linked to model-based reasoning. In particular, he has contributed to qualitative spatial reasoning, case-based reasoning (especially for design problems), constraint satisfaction for design and logistics problems, multi-agent systems, and intelligent user interfaces. His current work is oriented towards multi-agent systems and social computing, using concepts of game theory, constraint optimization and machine learning. In 1999, Professor Faltings co-founded Iconomic Systems, a company that developed a new agent-based paradigm for travel e-commerce. He has since co-founded 5 other startup companies and advised several others. Prof. Faltings has published more than 150 refereed papers on his work, and participates regularly in program committees of all major conferences in the field. He has served as associate editor of of the major journals, including the Journal of Artificial Intelligence Research (JAIR) and the Artificial Intelligence Journal. From 1996 to 1998, he served as head of the computer science department.

About this result

This page is automatically generated and may contain information that is not correct, complete, up-to-date, or relevant to your search query. The same applies to every other page on this website. Please make sure to verify the information with EPFL's official sources.

Related units (6)

Related units

Courses taught by this person (3)

Courses taught by this person

Related research domains (92)

A recommender system, or a recommendation system (sometimes replacing 'system' with a synonym such as platform or engine), is a subclass of information filtering system that provide suggestions for

Problem solving is the process of achieving a goal by overcoming obstacles, a frequent part of most activities. Problems in need of solutions range from simple personal tasks (e.g. how to turn on an a

In mathematics and computer science, an algorithm (ˈælɡərɪðəm) is a finite sequence of rigorous instructions, typically used to solve a class of specific problems or to perform a computation. Algo

Related research domains

Related publications (442)

Related publications

People doing similar research (103)

People doing similar research

CS-330: Artificial intelligence

Introduction aux techniques de l'Intelligence Artificielle, complémentée par des exercices de programmation qui montrent les algorithmes et des exemples de leur application à des problèmes pratiques.

CS-430: Intelligent agents

Software agents are widely used to control physical, economic and financial processes. The course presents practical methods for implementing software agents and multi-agent systems, supported by programming exercises, and the theoretical underpinnings including computational game theory.

ENG-704: EECS Seminar: Advanced Topics in Machine Learning

Students learn about advanced topics in machine learning, artificial intelligence, optimization, and data science. Students also learn to interact with scientific work, analyze and understand strengths and weaknesses of scientific arguments of both theoretical and experimental results.

A recommender system, or a recommendation system (sometimes replacing 'system' with a synonym such as platform or engine), is a subclass of information filtering system that provide suggestions for

A Practical Influence Approximation for Privacy-Preserving Data Filtering in Federated Learning

Panayiotis Danassis, Boi Faltings, Ljubomir Rokvic

Federated Learning by nature is susceptible to low-quality, corrupted, or even malicious data that can severely degrade the quality of the learned model. Traditional techniques for data valuation cannot be applied as the data is never revealed. We present a novel technique for filtering, and scoring data based on a practical influence approximation (`lazy' influence) that can be implemented in a privacy-preserving manner. Each participant uses his own data to evaluate the influence of another participant's batch, and reports to the center an obfuscated score using differential privacy. Our technique allows for highly effective filtering of corrupted data in a variety of applications. Importantly, we show that most of the corrupted data can be filtered out (recall of >90%, and even up to 100%), even under really strong privacy guarantees (ε≤1).

2023

Language Model Decoding as Likelihood–Utility Alignment

Boi Faltings, Valentin Hartmann, Martin Josifoski, Maxime Jean Julien Peyrard, Frano Rajic, Jiheng Wei, Robert West

A critical component of a successful language generation pipeline is the decoding algorithm. However, the general principles that should guide the choice of a decoding algorithm re- main unclear. Previous works only compare decoding algorithms in narrow scenarios, and their findings do not generalize across tasks. We argue that the misalignment between the model’s likelihood and the task-specific notion of utility is the key factor to understanding the effectiveness of decoding algorithms. To struc- ture the discussion, we introduce a taxonomy of misalignment mitigation strategies (MMSs), providing a unifying view of decoding as a tool for alignment. The MMS taxonomy groups decoding algorithms based on their implicit assumptions about likelihood–utility misalign- ment, yielding general statements about their applicability across tasks. Specifically, by an- alyzing the correlation between the likelihood and the utility of predictions across a diverse set of tasks, we provide empirical evidence supporting the proposed taxonomy and a set of principles to structure reasoning when choos- ing a decoding algorithm. Crucially, our analy- sis is the first to relate likelihood-based decod- ing algorithms with algorithms that rely on ex- ternal information, such as value-guided meth- ods and prompting, and covers the most di- verse set of tasks to date. Code, data, and models are available at https://github.com/epfl- dlab/understanding-decoding.

2023

ADER: Adaptively Distilled Exemplar Replay Towards Continual Learning for Session-based Recommendation

Boi Faltings, Xiaoyu Lin, Fei Mi

Session-based recommendation has received growing attention recently due to the increasing privacy concern. Despite the recent success of neural session-based recommenders, they are typically developed in an offline manner using a static dataset. However, recommendation requires continual adaptation to take into account new and obsolete items and users, and requires “continual learning” in real-life applications. In this case, the recommender is updated continually and periodically with new data that arrives in each update cycle, and the updated model needs to provide recommendations for user activities before the next model update. A major challenge for continual learning with neural models is catastrophic forgetting, in which a continually trained model forgets user preference patterns it has learned before. To deal with this challenge, we propose a method called Adaptively Distilled Exemplar Replay (ADER) by periodically replaying previous training samples (i.e., exemplars) to the current model with an adaptive distillation loss. Experiments are conducted based on the state-of-the-art SASRec model using two widely used datasets to benchmark ADER with several well-known continual learning techniques. We empirically demonstrate that ADER consistently outperforms other baselines, and it even outperforms the method using all historical data at every update cycle. This result reveals that ADER is a promising solution to mitigate the catastrophic forgetting issue towards building more realistic and scalable session-based recommenders.

ACM2022