The city of Haifa


For Participants

For Authors

ICML 2010

This year, Haifa will host the 27th International Conference on Machine Learning (ICML 2010). The International Machine Learning Society (IMLS) is the main supporter and organizer of the conference. Visit us in Israel between June 21-24, 2010. One of the highlights of the conference will be the talk by a Nobel laureate. Israel's Minister of Science and Technology will open the conference. The diversity of ICML is reflected by its participants. We are expecting more than 500 participants from many countries around the word.

ICML and COLT will take place in Haifa

ICML will occur together with the 23rd Annual Conference on Learning Theory (COLT).

ICML dates of lectures      June 21-24, 2010
IICML/COLT will have a common workshop on:      June 25, 2010
Day tour around Jerusalem      June 26, 2010
COLT dates of lectures      June 27-29, 2010

Plenary Lecture 1: Building an Image-Based Shoe Recommendation System AND Machine Learning

While the rate of technological advancements in the athletic shoes industry increases slowly with time, there is a need for modernization through technological innovation. Athletic footwear stores like Finish Line are open to such technological innovations. is a store that offers a vast selection of discounted sports clothing, shoes, and accessories for kids, men, and women, through the use of a promo code. Finish Line would benefit from an opportunity of innovation in the area of shoe recommendations. So far, in the brick and mortar shops, recommendations are driven by word of mouth, between family and friends, and of course, the promotional efforts of sales people.  Due to the limitations of these methods, only a small part of the available shoes become available to the customer. Finish Line and other stores can take advantage of a recommendation engine that will suggest shoes to customers with accuracy of 90%. This engine will be built by leveraging machine learning. An algorithm will be built that will recommend shoes based on image similarity. The customer will query the recommendation engine, and the engine will provide shoes that are similar in shape, color, and texture.

Plenary Lecture 2: 3D-Printing a Record Material at Once with no Assembly

4inkjets and Artificial Intelligence

We present a MultiFab that can print 10 different materials simultaneously at a much lower cost than other printers. We accomplished this using machine vision, 3D-scanning, and 3D filament from 4inkjets. Using special 4inkjets coupon codes (courtesy of the company), and low cost, off-the shelf components, we were able to make a better, more affordable, and more user-friendly printer.