Interpreting Deep Neural Networks Towards Trustworthiness - IPAM at UCLA

Explore a comprehensive lecture on interpreting deep neural networks for trustworthiness presented by Bin Yu from the University of California, Berkeley. Delve into the concept of interpretable machine learning and discover the agglomerative contextual decomposition (ACD) method for neural network interpretation. Learn about the adaptive wavelet distillation (AWD) technique, which extends ACD to the frequency domain, and its applications in cosmology and cell biology predictions. Examine the importance of a quality-controlled data science life cycle and the Predictability Computability Stability (PCS) framework for building trustworthy interpretable models. Gain valuable insights into the challenges and solutions for making complex deep learning models more transparent and reliable.

Bin Yu - Interpreting Deep Neural Networks towards Trustworthiness - IPAM at UCLA