Question 1650
{"accessible":false,"alternatives":[{"id":8211,"text":"errors based on noise and meaningless data brought into prediction or classification"},{"id":8212,"text":"estimate of how parameters should change in order to optimize the network "},{"id":8213,"text":"number of passes over the training dataset"},{"id":8214,"text":"number of times the algorithm's parameters are updated"},{"id":8215,"text":"number of training examples utilized in one iteration"}],"archived":false,"correctAlternativeId":8213,"explanation":"\u003cp\u003eAn\u0026nbsp;\u003cstrong\u003eepoch\u003c/strong\u003e\u0026nbsp;is a term used in\u0026nbsp;\u003ca href=\"https://radiopaedia.org/articles/machine-learning-1?lang=gb\"\u003emachine learning\u003c/a\u003e\u0026nbsp;and indicates the number of passes over the training dataset. Datasets are usually grouped into batches (especially when the amount of data is very large).\u003c/p\u003e\u003cul\u003e\u003cli\u003eoverfitting\u0026nbsp;is a problem in\u0026nbsp;machine learning\u0026nbsp;that introduces errors based on noise and meaningless data into prediction or classification\u003c/li\u003e\u003cli\u003ebackpropagation\u0026nbsp;is the process used to calculate the gradient associated with each parameter weighting to\u0026nbsp;estimate of how parameters should change in order to optimize the network\u0026nbsp;\u003c/li\u003e\u003cli\u003ebatch size\u0026nbsp;is a term used in\u0026nbsp;machine learning\u0026nbsp;and refers to the number of training examples utilized in one\u0026nbsp;iteration\u003c/li\u003e\u003cli\u003eiteration\u0026nbsp;is a term used in\u0026nbsp;machine learning\u0026nbsp;and indicates the number of times the algorithm's parameters are updated\u003c/li\u003e\u003c/ul\u003e","id":1650,"imageUrl":null,"imageAttribution":null,"imageAttributionCaseInfo":null,"firstQuestionPath":"/questions/1650","nextQuestionPath":null,"relatedArticles":[{"id":56142,"title":"Iteration (machine learning)","link":"/articles/iteration-machine-learning?lang=us"},{"id":56140,"title":"Batch size (machine learning)","link":"/articles/batch-size-machine-learning?lang=us"},{"id":56164,"title":"Backpropagation (machine learning)","link":"/articles/backpropagation-machine-learning?lang=us"},{"id":67692,"title":"Overfitting","link":"/articles/overfitting?lang=us"}],"alsoUsedIn":[{"id":236,"kind":"Course","title":"How to read a radiology AI paper","link":"https://radiopaedia.org/courses/how-to-read-a-radiology-ai-paper"},{"id":663,"kind":"Course","title":"Lectures by Jarrel Seah - page 663","link":"https://radiopaedia.org/courses/lectures-by-jarrel-seah/pages/663"},{"id":663,"kind":"Course","title":"Radiopaedia 2020 - Lectures Only - page 663","link":"https://radiopaedia.org/courses/radiopaedia-2020-exclusive/pages/663"},{"id":663,"kind":"Course","title":"2020 Virtual Conference Private Use - page 663","link":"https://radiopaedia.org/courses/2020-virtual-conference-private-use/pages/663"}],"stem":"\u003cp\u003eWhich best describes the meaning of the term 'epoch'\u0026nbsp;in machine learning?\u003c/p\u003e","menuLinks":[{"text":"Report problem with question","url":"https://docs.google.com/forms/d/e/1FAIpQLSfO3soWYhOjJ7yErSysyCe5V4A1CqW7WK3rDA7MtAkecMGqNw/viewform?entry.1624461248\u0026entry.553583435=https://radiopaedia.org/questions/1650"}],"attemptsPercentages":[{"alternativeId":"8211","percentage":14},{"alternativeId":"8213","percentage":45},{"alternativeId":"8212","percentage":13},{"alternativeId":"8214","percentage":17},{"alternativeId":"8215","percentage":11}],"promptToLogin":false,"questionManager":false,"articleId":"epoch-machine-learning"}