… To design output codes that are both discriminative and predictable, we propose a max-margin
formulation defined on the encoding transform. For each sample, the prediction from the …

… uses the maximum margins between classes as the criterion and constructs a bottom-up
binary tree based on the maximum margin. As a result, the corresponding coding matrix is more …

… to build the maximum margin tree. Concretely, for subclasses with the maximal maximum
margin, we … We can obtain a coding matrix M according to the maximum margin tree as follows: …

… where the support of sparse code is stable. Based on the derived margin, we propose a
hinge … Based on the local stability of sparse code support, we show in Sec. 2 that the decision …

… In this paper, we propose a lossless image coding scheme with discriminative structured
set prediction (SSP) model which incorporates max-margin Markov networks (M3Ns). Contrary …

… first attempt toward adapting maximum margin techniques for … Specifically, M3MDC maximizes
the margins between each … To maximize the margin between a pair of modeling outputs, …

… We present a simple geometric interpretation of these losses on the domain of margins. …
This a max-cut problem which we solve as an SDP. Note that both Schapire (1997); Li (2006) …

… the maximum-margin algorithm we propose has advantages in terms of accuracy and tunability
to specific loss functions. A maximum-margin … Pseudo-code of the algorithm is depicted in …

… maximum margin approach is common in ML, and naturally matches the communication
problem. Indeed, a NN decoder partitions the output … The pseudo code of our proposed …

In this paper, we present a new interpretation of AdaBoost.ECC and AdaBoost.OC. We
show that AdaBoost.ECC performs stage-wise functional gradient descent on a cost function, …