Vinay Sharma

Dilip U Shenoy, Vinay Sharma, Shiva HC Prasad

Most of the manufacturing firm aims to optimize their Supply Chain in terms of improved profitability of its products through value Addition. This study takes a critical look into the factors that affect the Performance of internal supply chain with respect to specific criteria. Accordingly, ranking these factors to get the critical dimensions of supply chain performance in the manufacturing industry. A semi-structured interview with the pre-defined set of questions used to collect the responses from decision makers of the firm. Multi criteria decision-making tool called TOPSIS is used to evaluate the responses and rank the factors. The results of this indicate that supplier relationship and inventory planning were most principal factors positively influencing on-time delivery of the product, production flexibility, cost savings, additional costs. This study helps to identify and optimize the process parameters using objective and subjective evaluation approach. The combined influence of the thought process of the manager to optimize the internal supply chain is extracted in this work.

Hadi Pouransari, Mojan Javaheripi, Vinay Sharma et al.

Knowledge distillation has been used to transfer knowledge learned by a sophisticated model (teacher) to a simpler model (student). This technique is widely used to compress model complexity. However, in most applications the compressed student model suffers from an accuracy gap with its teacher. We propose extracurricular learning, a novel knowledge distillation method, that bridges this gap by (1) modeling student and teacher output distributions; (2) sampling examples from an approximation to the underlying data distribution; and (3) matching student and teacher output distributions over this extended set including uncertain samples. We conduct rigorous evaluations on regression and classification tasks and show that compared to the standard knowledge distillation, extracurricular learning reduces the gap by 46% to 68%. This leads to major accuracy improvements compared to the empirical risk minimization-based training for various recent neural network architectures: 16% regression error reduction on the MPIIGaze dataset, +3.4% to +9.1% improvement in top-1 classification accuracy on the CIFAR100 dataset, and +2.9% top-1 improvement on the ImageNet dataset.