Recent and Upcoming Talks

Bayesian Optimization

Message Passing for State Estimation in Nonlinear Systems

Estimating the latent state of a dynamical system based on noisy observations is common challenge underlying many task in engineering, robotics, or climate science. Classical approaches to state estimation include Kalman filtering/smoothing, which is …

A Role for Message Passing in Data Assimilation?

Data-Efficient Machine Learning in Robotics

Data efficiency, i.e., learning from small datasets, is of practical importance in many real-world applications and decision-making systems. Data efficiency can be achieved in multiple ways, such as probabilistic modeling, where models and …

Data-Efficient Machine Learning in Robotics

Data efficiency, i.e., learning from small datasets, is of practical importance in many real-world applications and decision-making systems. Data efficiency can be achieved in multiple ways, such as probabilistic modeling, where models and …

Tutorial on Bayesian Optimization

Bayesian optimization is a useful tool for fast optimization of black-box functions. Typically, Bayesian optimization relies on Gaussian processes as a surrogate model for the unknown function, which can then be used to find a good trade-off between …

The Role of Uncertainty in Model-based Reinforcement Learning

Large-Scale Spatio-Temporal Inference via Message Passing

Estimating the latent state of a dynamical system based on noisy observations is a common challenge underlying many tasks in engineering, robotics, or climate science. Classical approaches to state estimation include Kalman filtering/smoothing, which …

Iterative State Estimation With Approximate EP

Estimating the latent state of a dynamical system based on noisy observations is a common challenge underlying many tasks in engineering, robotics, or climate science. Classical approaches to state estimation include Kalman filtering/smoothing, which …

Meta Learning via Bayesian Inference