Hi,
in your opinion, what are the best books to:
- start with learning in neural network area, and
- getting more in-depth knowledge in this domain?
Books/resources demanding mathematical prerequisites are ok.
Thanks
views:
991answers:
2
+1
A:
+4
A:
Neural networks have been obsoleted by more recent (and more successful) methods in machine learning: Support Vector Machines, Gaussian Processes, generative and descriptive models, and others. For the mathematically inclined reader, here are some good starting points:
- Hastie, Tibshirani and Friedman: The Elements of Statistical Learning: Data Mining, Inference, and Prediction
- Duda, Hart and Stork: Pattern Classification
- Christopher Bishop: Pattern Recognition and Machine Learning
- Rasmussen and Williams: Gaussian Processes for Machine Learning
Mind that these books are not tutorials, but graduate level books.
Thanks. How do you support the affirmation, "Neural networks have been obsoleted by ..." ? can you provide an objective explanation (or URL, or reference to an article etc) that states and explains this idea?
lmsasu
2009-03-25 19:30:29
"The Elements of Statistical Learning: Data Mining, Inference, and Prediction" - in my opinion, it is not good as a first reading. You can understand it it you already have a background in machine learning. Duda, Hart and Stork: Pattern Classification - a good book, better as "The elements" as a starting lecture, in my opinion. I do not know anything on thee last two books.
lmsasu
2010-06-21 17:31:21