Model compression as constrained optimization, with application to neural nets. Part I: general framework

Open in new window