Tariq Afzal

  • Citations Per Year
Learn More
Recently, DNN model compression based on network architecture design, e.g., SqueezeNet, attracted a lot of attention. Compared to well-known models, these extremely compact networks don’t show any accuracy drop on image classification. An emerging question, however, is whether these compression techniques hurt DNN’s learning ability other than classifying(More)
Published and projected benchmark numbers do not necessarily help predict the behavior of a user application on a particular microprocessor system. The real test comes when a user application is executed and timed on the microprocessor[5], which usually stresses the system in a different fashion than most benchmarks do. Unfortunately, the lack of stable(More)
  • 1