… Since we learn the surrogate loss with a much weaker constraint, our surrogate loss can …
outputs and metric values to train the surrogate loss once for all, without further fine-tuning. …

… This work aims to learn a prior-less and differentiable loss for surrogate regression through
a finitedimensional differentiable feature map ψ : Y → Rd. As a result, the novel framework …

… That is why we also refer to machine learning as learning from data. … loss for the surrogate
regression problem, then OKR and ILE framework coincide, with a kernel-induced loss …

… In self-supervised learning, a system is tasked with achieving a surrogate objective by
defining alternative targets on a set of unlabeled data. The aim is to build useful representations …

… between objects to improve the performance of listwise learning-to-rank algorithm. In this …
of loss function, the likelihood loss and cross entropy loss are defined as the surrogate loss …

This paper addresses the problem of recognizing human actions from RGB-D videos. A
discriminative relational feature learning method is proposed for fusing heterogeneous RGB and …

… learning and representing relations using embeddings [17, 31, 43, 8, 33, 1] and in developing
and injecting relational features into the learning … following, continuous surrogates are very …

… to improve relational topic models: 1) we relax the symmetric assumption and define
generalized relational topic … which is a good surrogate loss for the expected link prediction error, …

… The surrogate model acquires its general calibration capability from … 1, using relational-knowledge-transfer
learning, we divide the model training into two stages. The first-stage learning …

… We believe that REMAX performs better than GENE because it trains the VAE and the
surrogate model together in an end-to-end learning, while GENE separately trains VAE and KDE. …