Timezone: »
Operator-Valued Kernels (OVKs) and associated vector-valued Reproducing Kernel Hilbert Spaces provide an elegant way to extend scalar kernel methods when the output space is a Hilbert space. Although primarily used in finite dimension for problems like multi-task regression, the ability of this framework to deal with infinite dimensional output spaces unlocks many more applications, such as functional regression, structured output prediction, and structured data representation. However, these sophisticated schemes crucially rely on the kernel trick in the output space, so that most of previous works have focused on the square norm loss function, completely neglecting robustness issues that may arise in such surrogate problems. To overcome this limitation, this paper develops a duality approach that allows to solve OVK machines for a wide range of loss functions. The infinite dimensional Lagrange multipliers are handled through a Double Representer Theorem, and algorithms for \epsilon-insensitive losses and the Huber loss are thoroughly detailed. Robustness benefits are emphasized by a theoretical stability analysis, as well as empirical improvements on structured data applications.
Author Information
Pierre Laforgue (Télécom Paris)
Alex Lambert (Télécom ParisTech)
Luc Brogat-Motte (Télécom Paris)
Florence d'Alche-Buc (Télécom ParisTech, Université Paris-Saclay,Paris, France)
More from the Same Authors
-
2022 Poster: Learning to Predict Graphs with Fused Gromov-Wasserstein Barycenters »
Luc Brogat-Motte · Rémi Flamary · Celine Brouard · Juho Rousu · Florence d'Alché-Buc -
2022 Spotlight: Learning to Predict Graphs with Fused Gromov-Wasserstein Barycenters »
Luc Brogat-Motte · Rémi Flamary · Celine Brouard · Juho Rousu · Florence d'Alché-Buc -
2019 : Poster Session »
Ivana Balazevic · Minae Kwon · Benjamin Lengerich · Amir Asiaee · Alex Lambert · Wenyu Chen · Yiming Ding · Carlos Florensa · Joseph E Gaudio · Yesmina Jaafra · Boli Fang · Ruoxi Wang · Tian Li · SWAMINATHAN GURUMURTHY · Andy Yan · Kubra Cilingir · Vithursan (Vithu) Thangarasa · Alexander Li · Ryan Lowe -
2019 Poster: On Medians of (Randomized) Pairwise Means »
Stephan Clemencon · Pierre Laforgue · Patrice Bertail -
2019 Oral: On Medians of (Randomized) Pairwise Means »
Stephan Clemencon · Pierre Laforgue · Patrice Bertail -
2018 Poster: Structured Output Learning with Abstention: Application to Accurate Opinion Prediction »
Alexandre Garcia · Telecom-ParisTech Chloé Clavel · Slim Essid · Florence d'Alche-Buc -
2018 Oral: Structured Output Learning with Abstention: Application to Accurate Opinion Prediction »
Alexandre Garcia · Telecom-ParisTech Chloé Clavel · Slim Essid · Florence d'Alche-Buc