![Gabriel Peyré on Twitter: "The soft-argmax is the gradient of the soft-max ( log-sum-exp). Central to preform classification using logistic loss. Needs to be stabilized using the log-sum-exp trick. https://t.co/t2sANAWsLZ https://t.co/n0Jalhbm3d https ... Gabriel Peyré on Twitter: "The soft-argmax is the gradient of the soft-max ( log-sum-exp). Central to preform classification using logistic loss. Needs to be stabilized using the log-sum-exp trick. https://t.co/t2sANAWsLZ https://t.co/n0Jalhbm3d https ...](https://pbs.twimg.com/media/FKlmABrXsAMRfUW.jpg:large)
Gabriel Peyré on Twitter: "The soft-argmax is the gradient of the soft-max ( log-sum-exp). Central to preform classification using logistic loss. Needs to be stabilized using the log-sum-exp trick. https://t.co/t2sANAWsLZ https://t.co/n0Jalhbm3d https ...
![Convert the perspective function of log-sum-exp to cvx - CVX Forum: a community-driven support forum Convert the perspective function of log-sum-exp to cvx - CVX Forum: a community-driven support forum](http://ask.cvxr.com/uploads/default/original/1X/b23033b58ceb6bf3fda4d47a97e3c2b21204a41a.png)
Convert the perspective function of log-sum-exp to cvx - CVX Forum: a community-driven support forum
![information theory - Understanding the proof of the log sum inequality. - Mathematics Stack Exchange information theory - Understanding the proof of the log sum inequality. - Mathematics Stack Exchange](https://i.stack.imgur.com/y2Zxv.png)
information theory - Understanding the proof of the log sum inequality. - Mathematics Stack Exchange
![IJMS | Free Full-Text | Descriptor Selection via Log-Sum Regularization for the Biological Activities of Chemical Structure | HTML IJMS | Free Full-Text | Descriptor Selection via Log-Sum Regularization for the Biological Activities of Chemical Structure | HTML](https://www.mdpi.com/ijms/ijms-19-00030/article_deploy/html/images/ijms-19-00030-g002.png)