Recently I’ve started to see a lot of questions regarding to the “mathematics tutorials or supplementary materials for Machine Learning and AI” in the online discussions with the emergence of Stanford’s online AI and machine learning courses. As with the internet crowd, I’m going to participate these courses as well and I’ve always found the [...]
Archive for the ‘Machine Learning’ Category
Online Supplementary Mathematics Materials for Machine Learning and Artificial Intelligence Courses
04 Oct 2011 at 08:48
caglar
Artificial Intelligence, Computer Science, Engineering, Machine Learning, Mathematics, Web
Large Scale Bayesian Inference for Network Tomography
02 Jun 2011 at 21:41
caglar
Artificial Intelligence, Computer Science, Machine Learning, Mathematics, Network Engineering, statistics, Web
1 Introduction Major goal of network tomography is to infer the internal characteristics of network by only using data from the end nodes. Each node can either be a computer, router or a subnetwork. Broadly speaking large-scale network inference involves estimating network parameters (can be performance or other) based on traffic measurements at a limited subset of nodes [...]
OpenClassroom and E-learning
26 Mar 2011 at 20:46
caglar
Artificial Intelligence, Computer Science, Education, Machine Learning, Technology, Web
Recently I’ve discovered a very useful gem OpenClassroom from Stanford University. In that page there are lectures from several famous professors in Stanford University. The lectures are usually interactive and visual, for instance lecturer asks a question and the videos stops, then it continues after you’ve answered the question. I believe in free online education [...]
Importance Sampling
19 Jan 2011 at 16:24
caglar
Artificial Intelligence, Computer Science, Engineering, Machine Learning, Mathematics, Programming, science, statistics
Importance sampling is probably one of the easiest sampling algorithm and one of the most fundamental one as well. The main purpose of it is to estimate the properties of a particular distribution, while only having samples generated from a different distribution rather than the distribution of interest. Depending on the application, the term may [...]
Probability and Statistics Cheat sheet
Previously I’ve mentioned about a math/cs cheat sheet. It was pretty useful for me. But today I’ve found another interesting cheat sheet on the internet while searching for a definitely unrelated thing (convex optimization, if you really wonder and I found on John D. Cook‘s blog). This cheat sheet definitely extends the definition of a classical cheat [...]
Should implementing ML algorithms banned for Production Systems?
19 Dec 2010 at 17:51
caglar
Artificial Intelligence, Computer Science, Engineering, Machine Learning, Softwares
Nowadays everybody is talking about the how machine-learning algorithms can be useful your business, but now I’ll discuss here how it can harm your business . As a design principle(best practice), for the sake of security-preservation and efficiency in cryptographic systems, implementation of cryptographic algorithms isn’t recommended for production systems when there is already [...]
A reading list for Bayesian techniques
30 Nov 2010 at 10:32
caglar
Artificial Intelligence, Cognitive Science, Computer Science, literature, Machine Learning, science, statistics
Computational Cognitive Science Lab of Univ. of California Berkeley has a very nice reading-list for bayesian methods that are used in Machine-learning, statistics and Cognitive Science: A reading List for Bayesian Methods Related Posts:Rejection Letters from Peer ReviewersPapers that Everyone should read about Computer Science and Mathematics
Some great Free Books on Machine Learning
Here I list some of my favorite free (I suppose that not as in free as a free beer but free as freedom of speech) machine learning books: Bayesian Reasoning and Machine Learning By David Barber: This is a great free machine learning book for introduction. As you might notice it has a lot of [...]
A Conference on Transfer Learning by Learning Rich Generative Models
09 Oct 2010 at 20:06
caglar
Artificial Intelligence, Cognitive Science, Computer Science, Machine Learning
The study of transfer learning (or inductive transfer) is an important and popular subject in AI and ML. Because we as human beings can apply our knowledge in one subject to another subject quite easily. For instance we can recognize trucks with only the knowledge of cars. But current techniques for Statistical Learning have hard [...]
Bias and Variance Tradeoff
In simple terms, bias is the systematic difference(squared) between a quantity such as P(c|d)-the true conditional probability of data d being in class c- and a predicted value of this quantity (G(d)), averaged over training sets. Let’s say we have Y* which is the estimator of a parameter Y of a probability distribution. Y* is [...]
Recent Comments