In this paper we present an alternative strategy for fine-tuning the parameters of a network. We named the technique Gradual Tuning. Once trained on a first task, the network is fine-tuned on a second task by modifying a progressively larger set of the network’s parameters. We test Gradual Tuning on different transfer learning tasks, using networks of different sizes trained with different regu...

a facile and simple method was proposed to control the size and shape of the mgo nano structures with high surface area in the presence of efficient and cheap templates like peg 200, peg 600, peg 4000 and sorbitol at low temperature within a little time. nano rods and nanoparticles have been achieved by applying these templates and altering other growth parameters. the products were characteriz...

This paper addresses the model-based tuning of Proportional-Integral (PI) controllers focusing on the robustness/performance and servo/regulation trade-offs. First, a tuning rule is derived analytically by solving a simple H∞ weighted sensitivity problem, where the weight is chosen so that two design parameters permit to adjust the considered conflicting objectives. This way, the resulting tuni...

Proposed the PID controller parameters tuning method based-on New Luus-Jaakola (NLJ) algorithm and satisfaction idea. According to the different requirements of each performance index, designed the satisfaction function with fuzzy constraint attributes, and then determined the comprehensive satisfaction function for PID tuning by NLJ algorithm. Provided the steps of PID controller parameters tu...

Interconnected devices and intelligent applications have slashed human intervention in the Internet of Things (IoT), making it possible to accomplish tasks with less interaction. However, faces many problems, including lower capacity links, energy utilization, enhancement resources limited due its openness, heterogeneity, extensiveness. It is challenging route packets such a constrained environ...

Abstract. The choice of hyper-parameters (e.g. kernel parameters) can significantly affect generalization performance of large margin classifiers. In this paper we are concerned with the problem of tuning these values in the case of multi-class problems that have been recast into a set of binary problems. We report several experimental results comparing independent and joint tuning of the hyper...

Currently the parameters in a constraint solver are often selected by hand by experts in the field; these parameters might include the level of preprocessing to be used, the variable ordering heuristic or the suitable modelling approach. The efficient and automatic mechanism of parameters tuning for a constraint solver is a step towards making constraint programming a more widely accessible tec...

In this paper, for nonlinear, time-varying and delayed system, PID parameters self-tuning fuzzy control was applied to a level process. The method of using VC++ programming to realize fuzzy control algorithm and obtain the discrete fuzzy control table is proposed, which simplified the computing process greatly. Furthermore, combing the inquiry of fuzzy control table and PID function module in P...