Trapped in a Valley
Kyle and Linda discuss the challenges of gradient descent and how it can lead to getting trapped in local optima. They explore the concept of being stuck in a valley or hole, where the immediate surroundings seem optimal but may not be the best solution overall. They also touch on the importance of reassessing and adjusting step size during problem-solving.In this clip
From this podcast

Data Skeptic
[MINI] Gradient Descent
Related Questions