Lorenzo HERMEZ

Lorenzo HERMEZ

PhD Student at Telecom SudParis, Institut Polytechnique de Paris

Thesis' title: Multitask deep learning to predict the therapeutic impacts in neuro-functional rehabilitation

👋🏼 Hello there, I’m Lorenzo!

Illustration of combining vision and language modalities

👨🏻‍💻 I’m a final year PhD Student at Télécom SudParis - Institut Polytechnique de Paris.

🔬 My research focuses on machine learning and artificial intelligence for the characterization and evaluation of gait, as well as the prediction of therapeutic effects!

📚 I’m currently working towards the spatiotemporal characterization of gait through ML/IA methods.

📽️ I am also interested in helping others learn mathematics, through teaching and writing lessons/presentations.

🤖 Open Source Contributions

I strive to create clean and robust code suitable for open-source utilization. The software packages developed during my PhD research will be made available on GitHub, aiming to facilitate the application of angular kinematics data.

👨🏻‍🔬 Professional Experience

Research Scientist
Télécom SudParis - Institut Polytechnique de Paris (Actual)

  • My research focuses on the application of machine learning and artificial intelligence for the characterization and evaluation of gait patterns. I am particularly interested in predicting therapeutic effects using innovative methods that leverage non-time-normalized gait signals. By employing advanced algorithms and statistical techniques, I aim to enhance the understanding of human movement and contribute to the development of more effective rehabilitation strategies.

Lecturer (M1): Artificial Intelligence for Data Science
Télécom SudParis - Institut Polytechnique de Paris (Actual)

  • I taught an introductory course on machine learning, covering topics such as linear and logistic regressions, clustering, Gaussian Mixture Models (GMM), Support Vector Machines (SVM), and Principal Component Analysis (PCA). Additionally, I introduced deep learning concepts, including feedforward networks, gradient descent, and optimizers. The course included practical examples for quantified gait analysis.

Research Associate
Réseau de Transport d’Electricité (RTE) (February 2022 – August 2022)

  • During my tenure at RTE, I focused on characterizing electric consumption profiles by visualizing and characterizing data in an interpretable latent space using Keras. I also utilized Variational Autoencoders for reconstructing and generating time series data. Furthermore, I measured similarities and dissimilarities of consumption profiles in reduced dimensions.

Student Researcher
Télécom SudParis (October 2021 – January 2022)

  • As a research intern, I studied stochastic neural architectures for data regeneration, which included a comprehensive review of state-of-the-art methods. I also analyzed unsupervised and interpretable estimation methods and applied Variational Autoencoder architectures in my research.

Research Intern
Allianz France (January 2021 – June 2021)

  • At Allianz France, I worked on modeling and predicting temperatures across France. I visualized geographical data using Python (geojson) and utilized time series data from the pandas library for modeling and prediction. Additionally, I performed spatial interpolation to model temperature behavior at a national scale.

Throughout my academic and professional experiences, I have developed a strong foundation in artificial intelligence and data science. My roles have allowed me to teach fundamental concepts, conduct cutting-edge research, and apply advanced techniques to real-world problems. These experiences have equipped me with the skills to visualize and analyze complex data, develop innovative solutions, and contribute to both academic and industrial projects.

📚 Teaching

I will be uploading various lessons that I have created to the Teaching section of this website. These materials cover a range of topics in mathematics, statistics, and algorithms. If you have a particular interest in a specific subject within these fields, please do not hesitate to let me know. I am open to requests and can tailor the content to meet your needs. Whether you are looking for detailed explanations of mathematical concepts, statistical methods, or specific algorithms, I am here to assist you. Feel free to reach out with any inquiries or suggestions for future lessons.