SAP PhD CIFRE Advancing AI Ethics Fairness Explainability - F/M 

France, Ile-de-France 

951034847

Are you passionate about advancing responsible AI? ith AI evolving at an unprecedented pace, addressing the associated risks is more critical than ever, especially those stemming from bias and discrimination. These challenges not only compromise fairness but can also hinder system performance. Explainability techniques provide essential insights by revealing hidden biases in decision-making, laying the groundwork for effective interventions. This PhD position offers you the opportunity to tackle these challenges head-on, ensuring that AI systems operate reliably and equitably. In this role, you will:

• Develop advanced bias assessment techniques that leverage explainability to identify and understand bias sources.

• Design innovative methods that integrate explainability tools to measure bias throughout AI deployment and support targeted mitigation.

• Combine explainability and fairness metrics to generate actionable, data-driven insights for refining AI decision-making.

• Collaborate with experts in machine learning and responsible AI, leveraging state-of-the-art resources to bridge research with industry applications.

What you bring

• Graduate in MSc or Engineering degree in Data Science, Machine Learning, or a related field.

• Familiarity with model interpretability, causal inference, or counterfactual reasoning.

• Previous experience in AI explainability/fairness is advantageous.

• Demonstrated expertise in machine/deep learning.

• Proficiency in research methodologies: literature review, experimental design, and data analysis.

• Excellent Python programming skills, experience with deep learning frameworks (e.g. PyTorch, TensorFlow, Scikit-Learn).

• Knowledge of version control and containerization tools (e.g. Git, Docker).

• Strong analytical skills, creativity, passion for addressing societal challenges of AI.

• Fluent in English (wokring language), excellent written/verbal communication.

Contract: CIFRE

: 3 yearsStart date: as soon as possible

according to the conditions fixed by the French Minister of Higher Education and Research

Successful candidates might be required to undergo a background verification with an external vendor.

Limited Full Time