Almost all of the deployed applications of deep learning use supervised learning in which the machine is trained to predict human-provided annotations.
Then a support vector machine was trained using the nine dataset to correlate the catalyst preparation parameters and the kinetic parameters.
"If it produces the wrong answer," LeCun says, "you tell it the correct answer, and it adjusts the average". Once the machine is trained, it can begin to use this refined average to recognize what is contained in new images.
A relevance vector machine is trained over 3 dimensional feature vectors pertaining to global, local and rarity measures of conspicuity, to yield probabilistic values which form the saliency map.
The Eve machine is trained posing as either the Alice machine or the Bob machine.
This is traditionally achieved through a process called supervised learning, where the machine is trained on data that is a one-to-one match.
The Restricted Boltzmann Machine is trained in such a way that the obtained model parameter (varTheta) should minimize the negative log-likelihood of the given training data set.
Instead of spraying herbicide over an entire field and wasting most of it, this machine is trained to spray weeds directly, using 10% of the normal amount of herbicide.
So rather than hand-coding software routines with a specific set of instructions to accomplish a particular task, the machine is trained using large amounts of data and algorithms that give it the ability to learn how to perform the task.
A learning machine was trained using the Random Forest algorithm with prediction accuracy assessed using an out-of-bag estimation ("randomForest" package in R).
