Monte Carlo Dropout for Uncertainty Estimation in Neural Networks
Developed a Monte Carlo Dropout method to predict heat pump power usage using neural networks with quantified uncertainty in neural network predictions, enhancing model robustness and interpretability.
Designed and implemented a Monte Carlo Dropout model to predict heat pump power usage with neural network predictions. By applying dropout during inference, the model generates multiple stochastic forward passes, allowing for the quantification of predictive uncertainty. This approach enhances the robustness and interpretability of the model, providing valuable insights into the confidence of predictions and enabling more informed decision-making in energy management applications.
Professional Context
Tech Lead | VPP Systems Engineer at Bodil Energi
August 2024 – Present
Core Competences
Applied Skills
Distributed Energy Market Systems
Engineering Expertise
Forecasting and Statistical Modeling
Mathematical Foundation
Statistical Modeling
Implementation Stack
Python
TensorFlow / PyTorch