Sublime
An inspiration engine for ideas
Gary Marcus • Deep Learning Is Hitting a Wall
by distributing complex behavior across a large network of simple elements, the connections between which can change over time, we can accomplish virtually limitless tasks, all while continually learning new ones, even late in life.
Fei-Fei Li • The Worlds I See: Curiosity, Exploration, and Discovery at the Dawn of AI
Deep learning is brittle. As we saw in the opening chapter, deep learning can be perfect in one situation and utterly wrong in another.
Ernest Davis • Rebooting AI: Building Artificial Intelligence We Can Trust
guaranteed in “deeper” networks with more than two layers, it was possible to build a system that could produce results that were often good enough, opportunistically climbing up the mountain by taking small steps of the right sort, using a technique called backpropagation—now the workhorse of deep learning.*4 Backpropagation works by estimating th
... See moreErnest Davis • Rebooting AI: Building Artificial Intelligence We Can Trust
Machine Learning
Matt Mower • 1 card
Learning how to take actions within an environment to maximize reward, they realized, involved two related but potentially independent subproblems: action and estimation.