Generalization bounds and consistency for latent-structural probit and ramp loss

66 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

PAC-Bayesian Theory Meets Bayesian Inference (2016, neutral): “Theorem 1, due to Catoni [9], has been used to derive or study learning algorithms [12, 22, 34, 37].”
A New PAC-Bayesian Perspective on Domain Adaptation (2015, unknown): “…w ∈ R d : The taken approach is the one privileged in numerous PAC-Bayesian works ( e.g. , Langford & Shawe-Taylor, 2002; Ambroladze et al., 2006; McAllester & Keshet, 2011; Parrado-Hern´andez et al., 2012; Germain et al., 2009a; 2013), as it makes the risk of the linear classiﬁer h w and the…”
Training Deep Neural Networks via Direct Loss Minimization (2015, methodology): “In addition, for any ﬁnite (cid:15) , direct loss minimization is sub-gradient descent of some corresponding ramp loss (Keshet & McAllester, 2011).”
Advanced Structured Prediction (2014, background): “For the ramp loss, on the other hand, it can be shown (McAllester and Keshet, 2011) that when the norm of w goes to infinity, then the ramp loss approaches the risk.”
Object Detection with Grammar Models (2011, background): “Alternatively, taking Lmargin = L and Loutput = 0 gives the ramp loss that has been shown to be consistent in [14].”