Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!
Explore Geoff Hinton's groundbreaking concept of Capsule Networks in this 42-minute video presentation. Dive into an alternative approach to neural networks that divides layers into capsules, each responsible for detecting specific entities in input samples. Learn about the novel dynamic routing scheme that allocates information to higher-level capsules. Discover how Capsule Networks achieve state-of-the-art performance on MNIST and excel at recognizing highly overlapping digits. Understand the iterative routing-by-agreement mechanism and how it impacts the network's performance. Gain insights into this promising new direction in neural network architecture, presented by Yannic Kilcher based on the work of Sara Sabour, Nicholas Frosst, and Geoffrey E Hinton.