Votre mission
Project Summary
Recent years have witnessed a significant surge in retail investing, driven by several factors including the gamification of trading platforms, the rise of commission-free trading apps, and the adoption of frictionless, user-centric interfaces. Social media and « finfluencers » (financial influencers) have also played pivotal roles. While this has democratized access to financial markets, it has also correlated with a rise in problematic trading behaviors among non-professional investors. These behaviors, including overconfidence and emotionally-driven decision-making, often lead to adverse outcomes ranging from substantial financial loss to psychological distress, such as heightened stress, anxiety, and addictive patterns.
This project aims to address this challenge by developing an AI-driven system for the early detection and prediction of suboptimal trading behaviors. The core of this research involves modeling trader decision-making patterns by analyzing multimodal data streams. We will leverage a suite of physiological sensors to capture signals indicative of a trader’s cognitive and affective state—including eye-gaze dynamics, facial expressions, heart rate variability (HRV), and electrodermal activity (EDA) or galvanic skin response (GSR). These physiological data will be time-synchronized and fused with behavioral metrics captured from a bespoke simulated trading interface. The ultimate goal is to build robust predictive models that can identify precursors to poor decision-making, enabling potential interventions.
Role Description
We are seeking a highly motivated and talented PhD candidate to spearhead this cutting-edge research. The successful applicant will be responsible for the end-to-end execution of the project, from experimental design and data collection to the development and validation of novel machine learning models. This is a unique opportunity to work at the intersection of Artificial Intelligence, Human-Computer
Interaction (HCI), Affective Computing, and Finance, contributing to a field with significant real-world impact.
Vos responsabilités
Key Responsibilities
1. Trading Simulator Development: Design, implement, and iteratively improve a bespoke trading simulator for experimental data acquisition. This includes front-end interface development, back-end logic, and ensuring robust, high-precision logging of all user interactions.
2. Multimodal Sensing System Integration: Configure, integrate, and synchronize a multimodal physiological sensing system (e.g., eye-trackers, cameras, ECG/PPG, EDA/GSR) with the trading simulator.
3. Data Pipeline Management and Signal Processing: Design and execute the complete data pipeline, including data collection protocols, cleaning of noisy sensor data, preprocessing, and advanced feature engineering from high-dimensional physiological and behavioral time-series data.
4. Predictive Model Development: Develop, train, and validate machine learning models (e.g., using LSTMs, Transformers) for the detection and prediction of predefined trading behavior patterns.
Votre profil
Required (Must have):
· A Master of Science (MSc) degree in Computer Science, Data Science, Artificial Intelligence, Biomedical Engineering, Computational Neuroscience, or a related discipline.
· Strong proficiency in Python and its scientific computing ecosystem (e.g., NumPy, Pandas, Scikit-learn).
· Hands-on experience with at least one major deep learning framework: TensorFlow or PyTorch.
· A solid theoretical and practical foundation in machine learning, particularly with time-series data.
· A high degree of autonomy, a proactive attitude, and strong problem-solving skills.
· Excellent written and verbal communication skills in English.
Desired (Highly Valued and necessary for the project – can be learned after starting):
· Experience with physiological signal processing (e.g., filtering, artifact removal, feature extraction) for signals such as ECG/PPG, EDA, EEG, or eye-tracking data.
· Front-end or full-stack development experience (e.g., JavaScript frameworks, Dash/Streamlit).
Nice to have:
· Prior research experience in Human-Computer Interaction (HCI), Affective Computing, or Cognitive Science.
· Experience in designing and conducting human-subject experiments.
· A track record of scientific publications in relevant peer-reviewed conferences or journals.
Notre offre
1. A fully funded, full-time PhD position for 3 years.
2. A competitive salary in accordance with the University of Mons regulations.
3. Access to state-of-the-art lab facilities and high-performance computing resources.
4. A vibrant, collaborative, and interdisciplinary research environment.
5. Generous funding for attending international conferences and workshops.
6. Start Date: Fall 2025