Adrià Garriga-Alonso

Adrià Garriga-Alonso is a Research Scientist at FAR AI, where he focuses on advancing AI safety and interpretability. He began his role at FAR AI in March 2023, based in Berkeley, California. His research primarily involves understanding how neural networks function internally, with a particular interest in uncertainty quantification and the application of Bayesian principles in machine learning.13

Education and Background

Garriga-Alonso holds a PhD in Machine Learning from the University of Cambridge, where he was advised by Prof. Carl Rasmussen. His doctoral research emphasized improving uncertainty quantification in neural networks. He also earned a Master's degree in Computer Science from the University of Oxford and a Bachelor's degree in Computer Science from Universitat Pompeu Fabra in Barcelona.13

Previous Experience

Before joining FAR AI, he worked as a Member of Technical Staff at Redwood Research, focusing on interpretability research and software development. He also held research positions at Microsoft and the Music Technology Group at Universitat Pompeu Fabra, where he developed applications that integrate technology with music education.13

Research Contributions

Garriga-Alonso has contributed to various publications in the field of AI, particularly concerning the interpretability of neural networks and their alignment with human values. His work aims to ensure that AI systems are beneficial to society, exploring questions about how these systems acquire goals and agency.24

For more details about his work and contributions, you can view his LinkedIn profile here .