HiAI University of Haifa AI Conference
MAY 25-27, 2025
Workshops
27.5.2025 Tuesday
Carmel campus, Rabin building, Floor 5, Room number 5002
Gaussian Process Regression: Learning with Uncertainty
Mr. Nadav Cohen
09:00 - 11:00
Gaussian Process Regression (GPR) is a powerful probabilistic approach for modeling uncertainty in machine learning. It provides a flexible framework for making predictions while quantifying confidence, making it particularly useful in applications where uncertainty estimation is crucial.
In this workshop, we will cover the fundamental theory of GPR, exploring why it is such a valuable tool. We will provide motivation for key statistical concepts, including the role of priors, likelihoods, and posteriors, utilizing Bayes’ theorem to build intuition. In addition to the theoretical foundation, we will implement GPR in code and explore its applications in robotics. By the end of the session, participants will gain both conceptual understanding and practical experience with this versatile method.
11:00 - 11:30 - Coffee break
Physics-Informed Neural Networks (PINNs) - Part 1
Dr. Imri Aharoni
11:30 - 13:00
The workshop offers an introduction to Physics-Informed Neural Networks (PINNs), a cutting-edge machine learning approach that integrates fundamental physical principles directly into the model training process. By combining data-driven learning with physics-based constraints, PINNs offer generalization, data efficiency, and the ability to solve complex problems even where data is scarce or incomplete. PINNs are showing promising applications across diverse fields such as fluid dynamics, structural mechanics, robotics, climate science, and biological systems.
Participants will learn how to incorporate physical laws, such as partial differential equations, into neural network models and gain practical skills in building and training PINNs using PyTorch and DeepXDE.
The workshop is structured in two sessions-the first focuses on core concepts and methodologies, while the second offers a hands-on coding demonstration.
13:00 - 14:00 - Lunch break
Physics-Informed Neural Networks (PINNs) - Part 2
Dr. Imri Aharoni
14:00 - 15:30
HiAI University of Haifa AI Conference
MAY 25-27, 2025
Workshops
27.5.2025 Tuesday
Carmel campus, Rabin building
Floor 5, Room number 5002
Gaussian Process Regression: Learning with Uncertainty
Mr. Nadav Cohen
09:00 - 11:00
Gaussian Process Regression (GPR) is a powerful probabilistic approach for modeling uncertainty in machine learning. It provides a flexible framework for making predictions while quantifying confidence, making it particularly useful in applications where uncertainty estimation is crucial.
In this workshop, we will cover the fundamental theory of GPR, exploring why it is such a valuable tool. We will provide motivation for key statistical concepts, including the role of priors, likelihoods, and posteriors, utilizing Bayes’ theorem to build intuition. In addition to the theoretical foundation, we will implement GPR in code and explore its applications in robotics. By the end of the session, participants will gain both conceptual understanding and practical experience with this versatile method.
Mr. Nadav Choen
11:00 - 11:30
Coffee break
Physics-Informed Neural Networks (PINNs) - Part 1
Dr. Imri Aharoni
11:30 - 13:00
The workshop offers an introduction to Physics-Informed Neural Networks (PINNs), a cutting-edge machine learning approach that integrates fundamental physical principles directly into the model training process. By combining data-driven learning with physics-based constraints, PINNs offer generalization, data efficiency, and the ability to solve complex problems even where data is scarce or incomplete. PINNs are showing promising applications across diverse fields such as fluid dynamics, structural mechanics, robotics, climate science, and biological systems.
Participants will learn how to incorporate physical laws, such as partial differential equations, into neural network models and gain practical skills in building and training PINNs using PyTorch and DeepXDE.
The workshop is structured in two sessions-the first focuses on core concepts and methodologies, while the second offers a hands-on coding demonstration.
Dr. Imri Aharoni