I am wondering if there is any research out their about an kNN classifier with a optimized algorithm where a function is trained upon the training data set that maps a point to a value of k. Then, when the algorithm needs to classify a new point, it first looks for the nearest point in this trained function to find what value k it should use. Any thoughts or links to research like this?
We live in a world run by algorithms, computer programs that make decisions or solve problems for us. In this riveting, funny talk, Kevin Slavin shows how modern algorithms determine stock prices, espionage tactics, even the movies you watch. But, he asks: If we depend on complex algorithms to manage our daily decisions -- when do we start to lose control?
It is not necessary to learn programming to write algorithms. In fact we should learn to write algorithms way before learning to program. An algorithm is nothing but step by step solution to a problem. Each step should be an instruction for computer to execute. Instructions should be mentioned clearly without any ambiguity.
This is the second'I, Lawyer' podcast Artificial Lawyer/TromansConsulting has done with Sweden's leading legal tech writer, Fredrik Svärd, who runs the super-informative, Legaltech.se In this approximately 30 minutes chat we knock around a few subjects, such as where legal AI as an industry has got to; how the use of algorithms does not always mean there is any AI involved; why AI may be the answer to removing bias rather than the cause of it, and much, much more. We also give a special shout out to Lexpo, which is now just around the corner and will take place in Amsterdam 8 9 May 2017. Many thanks to Fred for organising and producing the podcast, which is below on Soundcloud.
Machine learning is the idea that there are generic algorithms that can tell you something interesting about a set of data without you having to write any custom code specific to the problem. Instead of writing code, you feed data to the generic algorithm and it builds its own logic based on the data. For example, one kind of algorithm is a classification algorithm. It can put data into different groups. The same classification algorithm used to recognize handwritten numbers could also be used to classify emails into spam and not-spam without changing a line of code.