A graphic explanation of the Bayes Theorem I enjoyed how the 3.16 section of the Stanford Artificial Intelligence class presented the Bayes theorem. Instead of giving a formula and expecting the alumni to apply it, they gave us a problem that the Bayes theorem would solve and expected, I believe, that we figured it out ourselves. Being as I am counting-challenged, it took me a while to figure out a way of solving it that was simple enough that I could be reasonably sure of my results. It turned out to be a very interesting detour. The problem was like this: the probability of having cancer is ; the probability of giving positive in a cancer test when you have cancer is ; and the probability of giving positive when you don't have cancer is . What is the probability of having cancer if you give positive in the test? It's interesting because it is sort of how it works. Let's assume a population of 1000: The bottom line represents the full population. Zooming in: In blue we have the people who don't have cancer but give positive.
Factor graph In probability theory and its applications, a factor graph is a particular type of graphical model, with applications in Bayesian inference, that enables efficient computation of marginal distributions through the sum-product algorithm. One of the important success stories of factor graphs and the sum-product algorithm is the decoding of capacity-approaching error-correcting codes, such as LDPC and turbo codes. A factor graph is an example of a hypergraph, in that an arrow (i.e., a factor node) can connect more than one (normal) node. When there are no free variables, the factor graph of a function f is equivalent to the constraint graph of f, which is an instance to a constraint satisfaction problem. Definition A factor graph is a bipartite graph representing the factorization of a function. where , the corresponding factor graph consists of variable vertices , and edges . and variable vertex when . , such as the marginal distributions. Examples An example factor graph is defined as
The Machine is Us/ing Us Final Version - Digital Ethnography Click here to download as a 3Mbps WMV file (55 MB) Click here to download in Quicktime format (96 MB) On January 31st I released the 2nd draft of The Machine is Us/ing Us hoping to receive feedback from my colleagues. (The first draft was only seen by my Digital Ethnography class 2 days before the 2nd draft was released on YouTube.) I sent it to 10 people. This new version is not a major revision. Big thanks to all those who sent me comments or posted their comments on YouTube, in the blogosphere, or on Mojiti. This work is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 License.
Amit’s Game Programming Information What’s on this page? I’m interested in producing complexity out of simple parts. This page contains bookmarks that I collected while working on games; I did not write most of the content linked from here. Determining how to move around on a map is an interesting problem. These pages are about specific techniques for pathfinding and object movement: My current favorite algorithm is A*, because it can handle varying terrain costs well, and it seems to be faster than most graph searching algorithms. Code and Demos A* for Beginners (with Basic code)A Java Applet demonstrating A* (mirror site) (be sure to use the Fudge method for best results)A* Explorer [Windows application] Lets you step through the A* algorithm.Flash pathfinding demo, includes source code.Python code for A* and other search algorithms — note that the astar_search function is only four lines long! Many times I play a game and wish that the computer opponents were written better. What techniques are useful in game AI? Notices
Data Mining: Finding Similar Items and Users Because we want to give kick-ass product recommendations. I'm showing you how to find related items based on a really simple formula. If you pay attention, this technique is used all over the web (like on Amazon) to personalize the user experience and increase conversion rates. To get one question out of the way: there are already many available libraries that do this, but as you'll see there are multiple ways of skinning the cat and you won't be able to pick the right one without understanding the process, at least intuitively. Defining the Problem To find similar items to a certain item, you've got to first define what it means for 2 items to be similar and this depends on the problem you're trying to solve: In each case you need a way to classify these items you're comparing, whether it is tags, or items purchased, or movies reviewed. Redefining the Problem in Terms of Geometry We'll be using my blog as sample. ["API", "Algorithms", "Amazon", "Android", "Books", "Browser"] That's 6 tags.
x is beginning of sweeping new online push Web Update MITx is beginning of sweeping new online push Editor’s Note: The Tech recognizes that the announcement of MITx constitutes a development that will affect students and faculty in important ways. Many of those voices are not yet reflected in this article. Please check back later in this week for additional MITx coverage, once Tech editors have vanquished their finals. MIT is developing an online educational platform that will be open-source, largely free, and let users outside of MIT earn certificates for completing Institute-caliber courses online. According to MIT Provost L. By doing “knowledge transfer” online through MITx, says Reif, “students come to a classroom or lab to do more of the enriching experiences they come to a campus for.” But MIT will offer the same online learning experience to the rest of the world as well, says Reif. MITx will serve two additional goals. What exactly will MITx be? MITx aims to combine the output of efforts like those in a single place.
Bucket - XKCD Wiki Bucket has an outer shell of metal; within the metal is a protective layer of high density plastic, in which may or may not reside pure HOH. There can only be speculation about what else the Bucket contains. Do not make our Bucket stupid or mean. Any stupiding of the Bucket will get you warned, kicked, and then banned.  Installing Download the source files from or using git, mirror the repository from here: $ wget $ wget $ wget $ wget Setup a database (MySQL recommended) - for example, on debian or ubuntu: $ sudo apt-get install mysql-server Create the tables described in bucket.sql. $ . People
What Can We Learn From Stanford University’s Free Online Computer Science Courses? At the end of July 2011, Stanford University announced that three introductory one-term undergraduate courses would be available free as online distance learning courses during the October to December 2011 term. Each course is taught be people who are leading figures in their fields, and in some case more-or-less the leading figures. Here are links to the descriptions of each of the courses: Machine Learning, taught by Professor Andrew Ng, Director of the Stanford Artificial Intelligence Lab, which is the main AI research organisation at Stanford University;Database Design, taught by Jennifer Widom, Professor and Chair of the Computer Science Department at Stanford University;Artificial Intelligence (AI), taught by Sebastian Thrun, Research Professor of Computer Science at Stanford University, and Peter Norvig, Director of Research at Google (who was a keynote speaker at the 2007 ALT Conference). 1. 2. 3. 4. 5-minute 2011 TED talk by Sebastian Thrun:
Intelligent Autonomous Systems - Home Teaching the On-Line Stanford class at UMass Lowell: Guest Post from Fred Martin December 16, 2011 at 6:58 am 160,000 Enroll Stanford’s Online AI Course—Is the University Obsolete? @aiclass: “Amazing we can probably offer a Master’s degree of Stanford quality for FREE. HOW COOL IS THAT?” Mark blogged about Stanford’s online Artificial Intelligence course in August. Sebastian Thrun and Peter Norvig’s online course, Introduction to Artificial Intelligence, was announced via email to a AAAI list early last summer. The course was advertised as equivalent to the Stanford University undergraduate AI course. @aiclass: “Advanced students will complete the same homeworks and exams as Stanford students. The course launched the first week of October. 160,000 students had signed up. At UMass Lowell I had taught my department’s AI course in Fall 2010. I told students they would be responsible for a final project on top of the Stanford requirements. I ended up with 16 students—12 grad and 4 undergrad. How It Works The lectures themselves are inspired by Khan Academy’s approach.
5 of the Best Free and Open Source Data Mining Software The process of extracting patterns from data is called data mining. It is recognized as an essential tool by modern business since it is able to convert data into business intelligence thus giving an informational edge. At present, it is widely used in profiling practices, like surveillance, marketing, scientific discovery, and fraud detection. There are four kinds of tasks that are normally involve in Data mining: * Classification - the task of generalizing familiar structure to employ to new data* Clustering - the task of finding groups and structures in the data that are in some way or another the same, without using noted structures in the data.* Association rule learning - Looks for relationships between variables.* Regression - Aims to find a function that models the data with the slightest error. For those of you who are looking for some data mining tools, here are five of the best open-source data mining software that you could get for free: Orange RapidMiner Weka JHepWork