Learning to rank with extremely randomized trees
نویسندگان
چکیده
In this paper, we report on our experiments on the Yahoo! Labs Learning to Rank challenge organized in the context of the 23rd International Conference of Machine Learning (ICML 2010). We competed in both the learning to rank and the transfer learning tracks of the challenge with several tree-based ensemble methods, including Tree Bagging (Breiman, 1996), Random Forests (Breiman, 2001), and Extremely Randomized Trees (Geurts et al., 2006). Our methods ranked 10th in the first track and 4th in the second track. Although not at the very top of the ranking, our results show that ensembles of randomized trees are quite competitive for the “learning to rank” problem. The paper also analyzes computing times of our algorithms and presents some post-challenge experiments with transfer learning methods.
منابع مشابه
Tree-Based Batch Mode Reinforcement Learning
Reinforcement learning aims to determine an optimal control policy from interaction with a system or from observations gathered from a system. In batch mode, it can be achieved by approximating the so-called Q-function based on a set of four-tuples (xt ,ut ,rt ,xt+1) where xt denotes the system state at time t, ut the control action taken, rt the instantaneous reward obtained and xt+1 the succe...
متن کاملCollisions Lead to Shallower Decision Trees
We study the following generalization of decision trees, which we call k-decision trees (where k > 1 is an integer). The algorithms are now allowed to query any arbitrary subset of the n input bits and the answer to the query tells if the number of ones in the subset is 0, 1, . . . , k − 1 or at least k. We also allow the algorithm to make similar queries for the number of zeroes. Our study of ...
متن کاملLearning to Branch in Mixed Integer Programming
The design of strategies for branching in Mixed Integer Programming (MIP) is guided by cycles of parameter tuning and offline experimentation on an extremely heterogeneous testbed, using the average performance. Once devised, these strategies (and their parameter settings) are essentially input-agnostic. To address these issues, we propose a machine learning (ML) framework for variable branchin...
متن کاملApplication of Additive Groves to the Learning to Rank Challenge
This is a description of the team AG submission to the Learning to Rank Challenge. This solution has scored 4th place in the main track. The primary algorithm used is Additive Groves of regression trees.
متن کاملTree-Based Ensemble Multi-Task Learning Method for Classification and Regression
Multi-task learning is an important area of machine learning that tries to learn multiple tasks simultaneously to improve the accuracy of each individual task. We propose a new tree-based ensemble multi-task learning method for classification and regression (MT-ExtraTrees), based on Extremely Randomized Trees. MTExtraTrees is able to share data between tasks minimizing negative transfer while k...
متن کامل