Learning without forgetting
When building a unified vision system or gradually adding new apabilities to a system, the
usual assumption is that training data for all tasks is always available. However, as the …
usual assumption is that training data for all tasks is always available. However, as the …
Borrowing treasures from the wealthy: Deep transfer learning through selective joint fine-tuning
Deep neural networks require a large amount of labeled training data during supervised
learning. However, collecting and labeling so much data might be infeasible in many cases …
learning. However, collecting and labeling so much data might be infeasible in many cases …
Yahoo! learning to rank challenge overview
O Chapelle, Y Chang - Proceedings of the learning to rank …, 2011 - proceedings.mlr.press
Learning to rank for information retrieval has gained a lot of interest in the recent years but
there is a lack for large real-world datasets to benchmark algorithms. That led us to publicly …
there is a lack for large real-world datasets to benchmark algorithms. That led us to publicly …
Multiple kernel learning-based transfer regression for electric load forecasting
Electric load forecasting, especially short-term load forecasting (STLF), is becoming more
and more important for power system operation. We propose to use multiple kernel learning …
and more important for power system operation. We propose to use multiple kernel learning …
Efficient and effective tree-based and neural learning to rank
As information retrieval researchers, we not only develop algorithmic solutions to hard
problems, but we also insist on a proper, multifaceted evaluation of ideas. The literature on …
problems, but we also insist on a proper, multifaceted evaluation of ideas. The literature on …
Boosting binary keypoint descriptors
T Trzcinski, M Christoudias, P Fua… - Proceedings of the …, 2013 - openaccess.thecvf.com
Binary keypoint descriptors provide an efficient alternative to their floating-point competitors
as they enable faster processing while requiring less memory. In this paper, we propose a …
as they enable faster processing while requiring less memory. In this paper, we propose a …
Gradient boosted feature selection
A feature selection algorithm should ideally satisfy four conditions: reliably extract relevant
features; be able to identify non-linear feature interactions; scale linearly with the number of …
features; be able to identify non-linear feature interactions; scale linearly with the number of …
A convex formulation for learning shared structures from multiple tasks
Multi-task learning (MTL) aims to improve generalization performance by learning multiple
related tasks simultaneously. In this paper, we consider the problem of learning shared …
related tasks simultaneously. In this paper, we consider the problem of learning shared …
Learning image descriptors with boosting
T Trzcinski, M Christoudias… - IEEE transactions on …, 2014 - ieeexplore.ieee.org
We propose a novel and general framework to learn compact but highly discriminative
floating-point and binary local feature descriptors. By leveraging the boosting-trick we first …
floating-point and binary local feature descriptors. By leveraging the boosting-trick we first …