Uploaded image for project: 'Flink'
  1. Flink
  2. FLINK-1807

Stochastic gradient descent optimizer for ML library

    XMLWordPrintableJSON

Details

    • Improvement
    • Status: Closed
    • Major
    • Resolution: Won't Do
    • None
    • None
    • Library / Machine Learning

    Description

      Stochastic gradient descent (SGD) is a widely used optimization technique in different ML algorithms. Thus, it would be helpful to provide a generalized SGD implementation which can be instantiated with the respective gradient computation. Such a building block would make the development of future algorithms easier.

      Attachments

        Issue Links

          is blocked by

          Improvement - An improvement or enhancement to an existing feature or task. FLINK-1901 Create sample operator for Dataset

          • Major - Major loss of function.
          • Resolved
          Is contained by

          New Feature - A new feature of the product, which has yet to be developed. FLINK-1889 Create optimization framework

          • Major - Major loss of function.
          • Closed
          is related to

          Improvement - An improvement or enhancement to an existing feature or task. FLINK-1979 Implement Loss Functions

          • Minor - Minor loss of function, or other problem where easy workaround is present.
          • Closed
          is required by

          Improvement - An improvement or enhancement to an existing feature or task. FLINK-1992 Add convergence criterion to SGD optimizer

          • Minor - Minor loss of function, or other problem where easy workaround is present.
          • Closed

          Improvement - An improvement or enhancement to an existing feature or task. FLINK-1993 Replace MultipleLinearRegression's custom SGD with optimization framework's SGD

          • Minor - Minor loss of function, or other problem where easy workaround is present.
          • Closed

          Improvement - An improvement or enhancement to an existing feature or task. FLINK-1994 Add different gain calculation schemes to SGD

          • Minor - Minor loss of function, or other problem where easy workaround is present.
          • Closed
          requires

          Improvement - An improvement or enhancement to an existing feature or task. FLINK-2396 Review the datasets of dynamic path and static path in iteration.

          • Not a Priority -
          • Open

          Activity

            People

              tvas Theodore Vasiloudis
              trohrmann Till Rohrmann
              Votes:
              0 Vote for this issue
              Watchers:
              5 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: