Dr. Julien Delaunay | Healthcare -Researcher in AI at Top Health Tech, Spain
Delaunay Julien is an accomplished researcher in the field of computer science, particularly known for his contributions to artificial intelligence (AI) and explainable machine learning models. His academic and professional journey spans multiple prestigious institutions across Europe and Canada. With a keen interest in making complex AI systems more interpretable, Julien has established himself as an expert in developing techniques that aim to enhance the transparency and trustworthiness of AI models, especially in natural language processing (NLP). His expertise is recognized globally through his publications, conference talks, and contributions to peer-reviewed journals. As an advocate for responsible AI, Julien actively engages in teaching, mentoring, and collaborative projects that bridge the gap between AI research and real-world applications. His passion for the field, combined with a commitment to advancing the next generation of AI researchers, makes him a leading figure in his area of study.
Profile:
Education:
Julien’s academic background reflects his dedication to mastering and advancing the field of computer science. He completed his Ph.D. in Computer Science from Inria and Rennes 1 University in France (2020–2023), with guidance from renowned advisors in the field. His research during this time focused on AI explainability, particularly the development of novel counterfactual explanation techniques. Prior to his Ph.D., Julien completed both his Master’s in Computer Science at Rennes 1 University and Sherbrooke University in Canada, where he specialized in artificial intelligence, data mining, and parallel programming. His undergraduate education at Rennes 1 University further solidified his foundation in object-oriented programming and web technologies. Through this extensive educational journey, Julien has developed a robust understanding of both the theoretical and practical aspects of computer science, making him an expert in machine learning and its interpretability.
Experience:
Julien has accumulated a wealth of professional experience in both academic and research settings, with key roles at prestigious institutions across Europe. His most recent position is as a researcher in computer science at the Top Doctors Group in Barcelona, Spain (2024–Present), where he continues his work on AI explainability. Prior to this, he held various academic positions, including graduate teaching assistant at Rennes 1 University and Institut Agro in France, and visiting researcher at Aalborg University in Denmark (2022–2023). His professional career began with multiple internships at Inria in Rennes, where he explored the intersection of machine learning and AI transparency. In addition to his academic roles, Julien has been actively involved in mentoring students and has presented at various international conferences, further establishing his reputation as a leader in the field.
Research Interest:
Julien’s research focuses on the development and enhancement of AI explainability techniques, particularly for complex machine learning models in natural language processing. He is passionate about making AI systems more transparent and understandable to both experts and non-experts alike. His work includes the development of novel counterfactual explanation techniques and model-agnostic frameworks that allow users to understand the decision-making processes of AI models better. Julien’s interests extend beyond technical innovations; he is deeply engaged in studying the impact of these explanation techniques on users’ trust and comprehension of AI. This research is crucial for promoting ethical AI and ensuring that AI models can be deployed responsibly in real-world applications. His ongoing work explores the adaptation of explanation methods for different user roles, ensuring that AI systems cater to diverse needs in various domains.
Award:
Delaunay Julien’s work has been recognized globally through various awards and nominations, reflecting the significance and impact of his contributions to AI research. His innovative approaches to explainability and interpretability have positioned him as a leading researcher in his field. His research has not only advanced the academic understanding of AI transparency but also contributed to practical applications, making him a strong candidate for the Best Researcher Award. His ability to develop new frameworks and methods in machine learning, alongside his academic leadership and outreach efforts, makes him an exemplary figure deserving of recognition. Julien’s commitment to improving AI accessibility, as well as his dedication to educating and mentoring future generations of researchers, strengthens his eligibility for prestigious awards in the domain of computer science and AI.
Publication:
Julien has authored and co-authored several notable publications, many of which have been well-received by the AI and machine learning research community. His works are frequently cited, contributing to the growing body of literature on AI transparency and explainability. Here are some of his key publications:
-
Delaunay J., Galárraga L., Largouët C. (2024). Does It Make Sense to Explain a Black Box With Another Black Box? Journal TAL: Explicabilité des modèles de TAL.
-
This paper compares and categorizes counterfactual explanation techniques for textual data, providing a deeper understanding of different counterfactual approaches.
📘 Cited by: 25+ articles
-
Delaunay J., Chaffin A. (2023). “Honey, Tell Me What’s Wrong”, Global Explainability of NLP Models through Cooperative Generation. Workshop on Analyzing and Interpreting Neural Networks for NLP at EMNLP 2023.
-
In this publication, Julien validated a novel global model-agnostic explanation technique for textual data.
📘 Cited by: 15+ articles
-
Delaunay J., Galárraga L., Largouët C., Van Berkel N. (2023). Adaptation of AI Explanations to Users’ Roles. Workshop on Human-Centered Explainable AI at CHI 2023.
-
This study explores how AI explanations can be adapted based on different user roles, contributing significantly to user-centered AI design.
📘 Cited by: 10+ articles
-
Wester J., Delaunay J., De Jong S., Van Berkel N. (2023). On Moral Manifestations in Large Language Models. Workshop on Moral Agent at CHI 2023.
-
The paper investigates the potential misinterpretation of large language models like ChatGPT as moral agents.
📘 Cited by: 8+ articles
-
Delaunay J., Galárraga L., Largouët C. (2022). When Should We Use Linear Explanations? International Conference on Information and Knowledge Management (CIKM).
-
This paper introduces the APE framework, which helps determine when linear explanations best approximate the decision boundaries of complex models.
📘 Cited by: 12+ articles
Conclusion:
Delaunay Julien’s work exemplifies innovation, leadership, and dedication in the field of computer science, specifically in AI explainability. His research has not only advanced theoretical knowledge but also addressed practical challenges in the deployment of AI systems. Julien’s ability to develop novel techniques that enhance the transparency of complex machine learning models has positioned him as a leading figure in his field. His numerous publications, contributions to major conferences, and mentorship roles highlight his influence in shaping the future of AI. Despite opportunities for growth in terms of broader public engagement and application-driven research, Julien’s contributions make him a highly deserving nominee for the Best Researcher Award. His career trajectory shows immense promise, and he is set to continue influencing the AI research landscape for years to come.