badmofo666

13 yıl öncediscuss

I think the best definition of intelligence is something like: "the ability of an agent to maximize its rewards by taking actions in some environment." In this sense, you could quantify intelligence by the agent's total reward over some time window, or how fast an agent learned the optimal strategy.

badmofo666

13 yıl öncediscuss

Sounds like what you're talking about is decision theory. Or, when mixed with learning: reinforcement learning. The monte carlo AIXI approximation would be an example. But, it really only works well on small toy problems (I.e. problems where the agent has only a small number of available actions it can perform).

badmofo666

13 yıl öncediscuss

I don't know much about literature, but I know philosophy is very important to the sciences. Especially the philosophy of science. Philosophy also deals with important things like logic, inference, etc. I took a couple philosophy courses, and was surprised how rigorous and technical they were.

badmofo666

13 yıl öncediscuss

All (good) statisticians validate their models. Nothing new there.

The main difference between these newer networks (besides much improved performance) is that the algorithms can handle "deeper" networks better (more hidden units). If we're talking about Deep Belief Networks, they're not much like the old ANNs. DBNs are generative probabilistic graphical models using Bayesian inference.

Conceptually, going deeper (LOL) allows the networks to learn higher level concepts. For example, a 1 layer ANN (perceptron) can only learn linear functions, while a deep network is able to internally form a belief of what, say, a cat is.

More technically: Much of the work in ML is deciding what your inputs (features) should be. When classifying text documents, should you use word counts, bag of words, word stemming, character counts, etc. Should the model be linear, polynomial, gaussian, trigometric, etc. Deep learners try to automatically do feature selection and control the degrees of freedom in the model for you.

Also, deep learning is catching on in some industries. It has recently had huge successes in speech recognition, and all major companies developing this technology have started using it (e.g. Siri for one).

comments