Learning multiple visual tasks while discovering their structure
By Carlo Ciliberto, Lorenzo Rosasco, Silvia Villa. 2015
17 citations recorded by the source corpus.
Multi-task learning is a natural approach for computer vision applications that require the simultaneous solution of several distinct but related problems, e.g. object detection, classification, tracking of multiple agents, or denoising, to name a few. The key idea is that exploring task relatedness (structure) can lead to improved performances. In this paper, we propose and study a novel sparse, nonparametric approach exploiting the theory of Reproducing Kernel Hilbert Spaces for vector-valued functions. We develop a suitable regularization framework which can be formulated as a convex optimization problem, and is provably solvable using an alternating minimization approach. Empirical tests show that the proposed method compares favorably to state of the art techniques and further allows to recover interpretable structures, a problem of interest in its own right.
Subfields: Artificial Intelligence, Computational Mechanics, Computer Vision and Pattern Recognition
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Qiang Yang
- Yu Zhang
- Carlo Ciliberto
- Giulia Denevi
- Dimitris Stamos
- Jianmin Wang
- Mingsheng Long
- Massimiliano Pontil
In Computational Mechanics
- Qiang Yang
- Yu Zhang
- Carlo Ciliberto
- Giulia Denevi
- Dimitris Stamos
- Massimiliano Pontil
- Dit-Yan Yeung
- Andreas Maurer
In Computer Vision and Pattern Recognition
- Qiang Yang
- Yu Zhang
- Massimiliano Pontil
- Dimitris Stamos
- Giulia Denevi
- Carlo Ciliberto
- Gabriella Csurka
- Dacheng Tao
Citations in the corpus
- A Survey on Multi-Task Learning (2017, background): “Problem (21) can also been interpreted from the perspective of reproducing kernel Hilbert spaces for vector-valued functions [85], [86], [87], [88].”
- Learning To Learn Around A Common Mean (2018, background): “The key difference is that in MTL the goal is to perform well on the observed tasks while in LTL the aim is to perform well on “future” tasks.”