Bookmark From <http://clopinet.com/isabelle/Projects/ICML2011/>
Machine Learning is the science of building hardware or software that can achieve tasks by learning from examples. The examples often come as {input, output} pairs. Given new inputs a trained machine can make predictions of the unknown output.- automatic reading of handwriting
- assisted medical diagnosis
- automatic text classification (classification of web pages; spam filtering)
- stock exchange predictions
We organize challenges to stimulate research in this field. The web sites of past challenges remain open for post-challenge submission as even-going benchmarks.
Feature selection (NIPS 2003): Book edited (CD with data and code)
Performance prediction (WCCI 2006):
Agnostic learning vs. prior knowledge (NIPS 2006 and IJCNN 2007): Learning causal dependencies (WCCI 2008 and NIPS 2008): Fast scoring in a large database (KDD cup 2009): KDD cup 2009 workshop page
Results
JMLR W&CP proceedings vol 7 Active Learning Challenge (AISTATS 2010 and WCCI 2010): Challenge website
AISTATS 2010 workshop
WCCI 2010 workshop
Machine Learning is the science of building hardware or software that can achieve tasks by learning from examples. The examples often come as {input, output} pairs. Given new inputs a trained machine can make predictions of the unknown output.
Examples of machine learning tasks include:
- automatic reading of handwriting
- assisted medical diagnosis
- automatic text classification (classification of web pages; spam filtering)
- stock exchange predictions
We organize challenges to stimulate research in this field. The web sites of past challenges remain open for post-challenge submission as even-going benchmarks.
Feature selection (NIPS 2003): Seventy five participants competed on five classification problems to make best predictions and select the smallest possible subset of relevant input variables (features). The tasks include: cancer diagnosis from mass-spectrometry data, handwritten digit recognition, text classification, and drug discovery. | Book edited (CD with data and code) |
Performance prediction (WCCI 2006): One hundred and forty-five five participants competed on five classification problems to make best predictions and predict their generalization performance on new unseen data. The tasks include: marketing, drug discovery, text classification, handwritten digit recognition, and ecology. | |
Agnostic learning vs. prior knowledge (NIPS 2006 and IJCNN 2007): This challenge has two tracks: the agnostic learning track and the prior knowledge track, corresponding to two versions of five datasets. The “agnostic track” version of the data is ready-to-use data preprocessed in a feature-based representation suitable for off-the-shelf machine learning packages. The identity of the features is not revealed. The “prior knowledge track” version of the data is just raw data, not always in a feature representation, coming with information about the nature and source of the data. Can you do better with the raw data and prior knowledge about the task? How far can you get with pure “black box learning”? | |
Learning causal dependencies (WCCI 2008 and NIPS 2008): What affects your health? What affects the economy? What affects climate changes? and… which actions will have beneficial effects? This series of competitions challenge the participants to discover the causes of given effects, based on observational data. The datasets include re-simulation data from models closely resembling real systems and real data for which the causal dependencies are known from experimental evidence. | |
| Fast scoring in a large database (KDD cup 2009): Customer Relationship Management (CRM) is a key element of modern marketing strategies. The KDD Cup 2009 offered the opportunity to work on large marketing databases from the French Telecom company Orange to predict the propensity of customers to switch provider (churn), buy new products or services (appetency), or buy upgrades or add-ons proposed to them to make the sale more profitable (up-selling). | KDD cup 2009 workshop page Results JMLR W&CP proceedings vol 7 |
| Active Learning Challenge (AISTATS 2010 and WCCI 2010): Labeling data is expensive, but large amounts of unlabeled data are available at low cost. Such problems might be tackled from different angles: learning from unlabeled data or active learning. In the former case, the algorithms must satisfy themselves with the limited amount of labeled data and capitalize on the unlabeled data with semi-supervised learning methods. In the latter case, the algorithms may place a limited number of queries to get labels. The goal in that case is tooptimize the queries to label data and the problem is referred to as active learning. | Challenge website AISTATS 2010 workshop WCCI 2010 workshop |
No comments:
Post a Comment