For those considering an autodidactic alternative, this is for you. You can't go deeply into every machine learning topic. There's too much to learn, and the field is advancing rapidly. Motivation is far more important than micro-optimizing a learning strategy for some long-term academic or career goal. If you're trying to force yourself forward, you'll slow down.
This is the shortest path I see towards machine intelligence: first, we develop ways to allow specialized AIs to manipulate formal concepts, write programs, run experiments, and at the same time develop mathematical intuition (even creativity) about the concepts they are manipulating. Then, we use our findings to develop an AI scientist that would assist us in AI research, as well as other fields. It would be a specialized superhuman artificial intelligence to be applied to scientific research. This would tremendously speed up the development of AI. At first we would apply it to solve well-scoped problems: for instance, developing agents to solve increasingly complex and open-ended games.
An artificial intelligence revolution has been eagerly awaited since the late 1950s, when pioneering IBM researcher Arthur Samuel trained the world's first self-learning computer to play a mean game of checkers. But only in the past few years has the long-promised technology become mature, effective and -- thanks to a variety of new offerings -- readily accessible to the channel. AI and machine learning are now taking the industry by storm, with the cloud fast-tracking adoption of solutions that make decisions, automate business processes, deliver predictions and insights and learn from their own experience. Next-gen startups are at the forefront of the revolution, delivering infrastructure, development frameworks, and intelligent applications that allow enterprises to take advantage of their data in ways never before possible.