This March, I had the privilege of attending the Machine Learning Summer School (MLSS) at the Okinawa Institute of Science and Technology (OIST) in Japan. The summer school covers a wide range of topics, from foundational principles to the latest advancements in machine learning, all delivered by leading experts in the field.
Each day consisted of multiple lectures on various aspects of machine learning. One of my favourite lectures included “What Can We Further Learn from the Brain for AI?” by Kenjo Doya. The presentation highlighted how AI has benefited from brain-inspired approaches, such as deep learning, and suggested that further advancements in AI can be achieved by continuing to draw insights from brain science. However, Doya also emphasized that both engineering intelligent systems without biological constraints and reverse engineering the brain’s existing intelligence mechanisms are important research directions. One of the most interesting parts of the lecture was how Doya’s team found that certain neurotransmitters regulate behaviour in ways analogous to how certain parameters function in reinforcement learning.
Poster sessions held during dinners provided opportunities for discussion and networking with both summer school attendees and experts. During one of these sessions, I presented my poster on learning reward functions from videos. This was a useful opportunity for me to learn to explain my research to people in other fields of machine learning, and for meeting people to collaborate with in the future.
During our spare time, we explored Okinawa; a beautiful tropical island with a fascinating history, rich culture, and abundant wildlife. The island offered stunning beaches, delicious local cuisine, and unique cultural sites. Exploring the natural beauty and vibrant traditions of Okinawa added an unforgettable dimension to the overall experience.
Thank you to Emlyn for taking the time to share his experiences of this fantastic sounding Summer School.
If you have something you want to share with the AgriFoRwArdS community, please get in touch at agriforwards.cdt@lincoln.ac.uk.
Be First to Comment