Hacker News Comments on
The Elements of Statistical Learning: Data Mining, Inference, and Prediction, Second Edition (Springer Series in Statistics)
·
5
HN comments
- This course is unranked · view top recommended courses
Hacker News Stories and Comments
All the comments and stories posted to Hacker News that reference this book.How does this (or its EdX version) compare with ESL? (https://amzn.com/0387848576)
⬐ whymauriI took this class in-person before reading ESL. I'd say there's more overlap between this class (18.650) and the class textbook All of Statistics (Wasserman) than ESL.That said, ESL is a better companion than Wasserman if you want to apply the statistics to ML and don't plan on studying the graduate-level statistics courses. ESL + 18.650 + 9.520 (Statistical Learning Theory, Poggio and Sasha Raklin) covers 95% of the math and statistics I've seen in ML research.
⬐ scared2Link to 9.520 https://cbmm.mit.edu/lh-9-520
Some of the best textbooks in statistics and machine learning:Applied -------
Hosmer et al., Applied Logistic Regression. An exhaustive guide to the perils and pitfalls of logistic regression. Logistic regression is the power tool of interpretable statistical models, but if you don't understand it, it will take your foot off (concretely, your inferences will be wrong and your peers will laugh at you.) This book is essential. Graduate level, or perhaps advanced undergraduate, intended for STEM and social science grad students.
https://www.amazon.com/Applied-Logistic-Regression-Probabili...
Peter Christen's Data Matching. Record Linkage is a relatively niche concept, so Christen's book has no right to be as good as it is. But it covers every relevant topic in a clear, even-handed way. If you are working on a record linkage system, then there's nothing in this book you can afford not to know. Undergraduate level, but intended for industry practitioners.
https://www.amazon.com/Data-Matching-Techniques-Data-Centric...
Max Kuhn's Applied Predictive Modeling. Even if you don't use R, this is an incredibly good introduction to how predictive modeling is done in practice. Early undergraduate level.
http://appliedpredictivemodeling.com/
Theoretical -----------
The Elements of Statistical Learning. Probably the single most respected book in machine learning. Exhaustive and essential. Advanced undergraduate level.
https://www.amazon.com/Elements-Statistical-Learning-Predict...
Kevin Murphy's Machine Learning: A Probabilistic Perspective. Covers lots of the same ground as Elements but is a little easier. Undergraduate level.
https://www.amazon.com/Machine-Learning-Probabilistic-Perspe...
Taboga's Lectures on Probability Theory and Mathematical Statistics. Has the distinction of being available for free in web-friendly format at https://www.statlect.com/ . Undergraduate level for math majors.
https://www.amazon.com/Lectures-Probability-Theory-Mathemati...
> In Searle’s time, the dominant AI paradigm was GOFAI (Good Old-Fashioned Artificial Intelligence.)Russel and Norvig's book is probably the best introduction to "old fashioned" AI:
https://www.amazon.com/Artificial-Intelligence-Modern-Approa...
GOFAI may not have lead directly to true AI, but it produced a ton of useful algorithms such as A* and minimax. Although the attention has turned to machine learning algorithms (à la https://www.amazon.com/Elements-Statistical-Learning-Predict...) the hybrid of GOFAI and ML has produced some extraordinary results, such as AlphaZero:
Thanks for the link.https://www.amazon.com/Elements-Statistical-Learning-Predict...
vs
https://www.amazon.com/Introduction-Statistical-Learning-App...
The former seems to assume prior knowledge and the latter seems more beginner friendly, except it uses R (in case anyone wants the distinction).
⬐ disgruntledphd2R is well worth learning if you're going to be doing any serious work with data. It's a weird-ass language, but it does data analysis/graphing/modelling really, really well.⬐ gubbeThese books assume some degree of Linear Algebra, ESL is definitely not for a entry level.I would like to add www.codingthe matrix.com for any developers, which has a MOOC as well (Coursera), as a complement to Strang.
These are both good options:http://www.amazon.com/Pattern-Recognition-Learning-Informati... http://www.amazon.com/The-Elements-Statistical-Learning-Pred...