Many problems in classical machine learning can be cast as a constrained or unconstrained optimization problem. There is a vast variety of optimization problems from classical machine learning and hence, hundreds and thousands of solvers have been implemented.

Here, we introduce GENO (GENeric Optimization), a framework that lets the user specify a constrained or unconstrained optimization problem from classical machine learning in an easy-to-read modeling language. Then, a solver is generated automatically, i.e., Python code, that can solve this class of optimization problems. The generated solver is usually as fast as hand-written, problem-specific, and well-engineered solvers. Often the solvers generated by GENO are faster by a large margin compared to recently developed solvers that are tailored to a specific problem class.

GENO is the first framework for classical machine learning problems that is flexible and at the same time efficient.

An online interface to our framework can be found at http://www.geno-project.org.