News
-
PhD position on machine learning for epilepsy detection in ULB, Brussels, Belgium
The Laboratory of Translational Neuroanatomy and Neuroimaging (LN2T, https://ln2t.ulb.be/) and the Machine Learning Group (MLG, mlg.ulb.ac.be)
of the Université libre de Bruxelles (ULB) are looking for a talented and motivated PhD student.
Candidates are invited to apply for a fully funded PhD position that will contribute to the design and assessment of novel machine learning techniques for the automatic detection of epileptiform discharges in magnetoencephalography (MEG) data. The successful candidate will benefit of a large database of MEG data acquired in epileptic patients and healthy subjects to develop machine learning models aiming at the automatic detection and localization of interictal epileptiform discharges. This position offers a unique opportunity to develop an interdisciplinary expertise in the fields of machine learning, MEG signal processing, and clinical MEG.
The position is funded for 4 years thanks to a research grant of the Fonds Erasme (Brussels, Belgium) https://fondserasme.org/fr
Expected starting date: 1st October 2024.
The successful candidate will work in a collaborative environment gathering two laboratories (LN2T and MLG) that are joining their expertise to conduct the proposed project. The LN2T is a multidisciplinary team composed of physicists, engineers, neuropsychologists, and neurologists. It holds a multimodal functional neuroimaging platform comprising a whole-scalp cryogenic MEG (Triux, MEGIN), an OPM-MEG facility, a MEG-compatible high-density EEG (Geodesic), and a hybrid PET-MRI system (3T MRI, Signa GE Healthcare). The ULB Machine Learning Group carries out theoretical research on statistical machine learning, time series forecasting and predictive modeling, and takes part to interdisciplinary research projects at international and national levels across diverse domains. In neuroscience, the MLG developed state-of-the-art methods in mental state classification, electrophysiological diagnostic tools (EDT) and prosthesis control. MLG is involved in several research initiatives at international and national levels: it is a member of the CLAIRE confederation, a member of the TRAIL Walloon initiative on AI and of FARI, the Brussels AI institute.
Both labs (LN2T and MLG) are on the ULB campuses in Brussels with easy access by public transports.
Candidates must hold a master’s degree (Bologna process) in engineering, computer science, physics, or a related discipline. A strong desire to learn about neuroscience and work in a multidisciplinary research environment is required. Strong proficiency in statistical programming languages (preferably Python and Matlab) is desirable. Preliminary experience in the field of machine learning or MEG/EEG signal processing is desirable but not required. French is not mandatory for the position.
Salary will correspond to the PhD grant of the ULB (around 2250-2400€/month for the PhD (net salary)) and should allow a comfortable life considering the living cost in Brussels.
Informal enquiries and applications can be addressed to Xavier De Tiège at xavier.de.tiege@ulb.be and Gianluca Bontempi at gianluca.bontempi@ulb.be.
Applications should comprise a detailed curriculum vitae (including programming skills), a letter of motivation, and the names of 2 reference persons.
Closing date: As soon as the position is filled.
-
“Causal Machine Learning for Business Decision-Making” by dr. Wouter Verbeke
March 30th 2023, 12:15, room S.R42.4.502 (Solbosch)
Machine learning is a powerful tool to support business decision-making. For instance, predictive models can be learned from data to anticipate the future and to make informed decisions, with the eventual objective of optimizing the efficiency and effectiveness of business operations. Even better than having predictive models, which tell you what will happen, is to have prescriptive models, which tell you what to do so as to optimize the outcome of interest. To this end, in the field of prescriptive analytics and operations research, simulation models are developed by an human expert modeler in the form of a series of mathematical equations. As an alternative approach, causal machine learning can be adopted to learn to predict the future as a function of the decisions that are made. In other words, causal machine learning models estimate the net effect on the outcome(s) of interest that would be caused by various potential business decisions. As such, these models may directly indicate the optimal decision. In this talk, I will demonstrate the use and need for causal machine learning by discussing on a number business cases. I will discuss on the challenges in estimating causal effects and learning a simulation model from data, and introduce some basic causal machine learning methods.
For more information, please contact John Iacono (john.iacono@ulb.be)
-
Seminar: “Conformal Prediction & Complex Data Analytics” by Dr. Matteo Fontana
March the 23th, 2023 at 3:00PM, on Teams: link.
In the latest years, scholars started focusing on how to develop statistical tool for the analysis of population of complex data, such as high-dimensional vector data and functions, but also more complex data objects such as functional time series or sets of graphs, either labelled or unlabelled. The present works adds to this literature by focusing on a strangely overlooked area, namely the formulation of prediction sets. By exploiting cutting edge techniques in the realm of machine learning, we propose a very powerful forecasting methodology, able to identify prediction regions in a very general sense, applicable to the great variety of possible data object the modern data scientist has to analyse. Our method, strongly based on Conformal Prediction, is model-free, achieves finite-sample validity, is computationally efficient and it identifies interpretable prediction sets, in the shape of a parallelotope. In the talk I will briefly present the basic ideas behind the methodology, its implementation to the functional and graph case, both labelled and unlabelled, as well as applications on real world data.
Matteo Fontana is a Project Officer at the Joint Research Centre of the European Commission, where is part of the Centre of Advanced Studies project “Computational Social Science for Policy”. From 2019 to 2021 he has been a Postdoctoral Researcher at the Modelling and Scientific Computing Lab of Politecnico di Milano, where he was involved in the development of an early warning system for geo-hazards in collaboration with the Italian Space Agency. He holds a PhD in Management Engineering from Politecnico di Milano, where he studied the application of novel statistical learning methodology to climate change economics research. He is a statistician/data scientist by training, he is mainly interested in applications of data science in economics, demography and migration studies. His main theoretical interests lie in the realm of nonparametric statistics (namely hypothesis testing and forecasting), as well as in the modelling of complex data objects.
-
MLG Researchers Launch “Be The Expert”, an experiment on human decision-making
While the promise of more knowledge makes the use of expert groups appealing, reality suggests that making good use of a larger number of experts is not so simple. Beliefs held by individuals tend to spread, which in turn makes decision-making by collectives less reliable.
The main goal of this experiment is to shed light on how AI can help us make better decisions by reducing the effect of our biases.
Can we use algorithms to reduce bias in decision-making? Your contributions to our data collection will allow us to show whether this could be true!
Complete the Be The Expert test now and discover if the AI finds bias in your decision-making.
-
MLG/AI-lab researcher obtains best poster award AI for behaviour change workshop
Elias Fernandez and co-authors were awarded the best poster award at the AAAI 2021 conference workshop on AI for Behavior Change. The poster was titled Delegation to autonomous agents promotes cooperation on collective-risk dilemmas. You can find the abstract below. This research is a collaboration with the VUB AI lab and GAIPS in Portugal, while being supported by a NESTA collective intelligence grant.
Abstract > Home assistant chat-bots, self-driving cars, drones or automated negotiations are some of the several examples of autonomous (artificial) agents that have pervaded our society. These agents enable the automation of multiple tasks, saving time and (human) effort. However, their presence in social settings raises the need for a better understanding of their effect on social interactions and how they may be used to enhance cooperation towards the public good, instead of hindering it. To this end, we present an experimental study of human delegation to autonomous agents and hybrid human-agent interactions centered on a public goods dilemma shaped by a collective risk. Our aim to understand experimentally whether the presence of autonomous agents has a positive or negative impact on social behavior, fairness and cooperation in such a dilemma. Our results show that cooperation increases when participants delegate their actions to an artificial agent that plays on their behalf. Yet, this positive effect is reduced when humans interact in hybrid human-agent groups. Finally, we show that humans are biased towards agent behavior, assuming that they will contribute less to the collective effort.
-
MLG researchers obtain Best Paper Award at ICCCI 2020
Axel Abels and Tom Lenaerts were, together with their collaborators, awarded with the Best Paper award at the International Computational Collective Intelligence (ICCCI) for the paper entitled “How expert confidence can improve collective decision-making in contextual multi-armed bandit problems. Below the abstract of the paper. This work was supported by a FRIA grant awarded by Fondation de la Recherche Scientifique (F.R.S.-FNRS) of the Walloon-Brussels Federation.
In collective decision-making (CDM) a group of experts with a shared set of values and a common goal must combine their knowledge to make a collectively optimal decision. Whereas existing research on CDM primarily focuses on making binary decisions, we focus here on CDM applied to solving contextual multi-armed bandit (CMAB) prob- lems, where the goal is to exploit contextual information to select the best arm among a set. To address the limiting assumptions of prior work, we introduce confidence estimates and propose a novel approach to de- ciding with expert advice which can take advantage of these estimates. We further show that, when confidence estimates are imperfect, the pro- posed approach is more robust than the classical confidence-weighted majority vote.
-
Combining people and machines
NESTA UK reporting on the behavioural experiments we performed in collaboration with the VUB AI lab and Francisco C. Santos team in INESC-ID in Lisbon.
Want to know more? Follow this link to get information and download the full report
We tested whether using AI improves co-operation and decision-making in collective risk situations. In such situations, the group must coordinate to resolve a common challenge, such as the climate crisis, to avoid detrimental outcomes for everyone. The experiment results show that delegating responsibility for actions to autonomous agents increased co-operation and co-ordination within the group.
