{"title":"Appendix B. Probability","authors":"","doi":"10.1515/9780804781640-015","DOIUrl":"https://doi.org/10.1515/9780804781640-015","url":null,"abstract":"","PeriodicalId":430009,"journal":{"name":"Pricing and Revenue Optimization","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124549605","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Chapter 12. Pricing and Revenue Optimization and Customer Acceptance","authors":"","doi":"10.1515/9780804781640-013","DOIUrl":"https://doi.org/10.1515/9780804781640-013","url":null,"abstract":"","PeriodicalId":430009,"journal":{"name":"Pricing and Revenue Optimization","volume":"31 3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123423298","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Appendix B probability","authors":"A. Černý","doi":"10.1515/9781400831487-016","DOIUrl":"https://doi.org/10.1515/9781400831487-016","url":null,"abstract":"","PeriodicalId":430009,"journal":{"name":"Pricing and Revenue Optimization","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-12-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121601366","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"5 Optimization","authors":"","doi":"10.1515/9781503614260-006","DOIUrl":"https://doi.org/10.1515/9781503614260-006","url":null,"abstract":"Once you've gathered your data, selected a representation to work with, chosen a framework for function approximation, specified an error metric, and expressed your prior beliefs about the model, then comes the essential step of choosing the best parameters. If they enter linearly, the best global values can be found in one step with a singular value decomposition, but as we saw in the last chapter coping with the curse of dimensionality can require parameters to be inside nonlinearities. This entails an iterative search starting from an initial guess. Such exploration is similar to the challenge faced by a mountaineer in picking a route up a demanding peak, but with two essential complications: the search might be in a 200-dimensional space instead of just 2D, and because of the cost of function evaluation you must do the equivalent of climbing while looking down at your feet, using only information available in a local neighborhood. The need to search for parameters to make a function extremal occurs in many kinds of optimization. We already saw one nice way to do nonlinear search, the Levenberg– Marquardt method (Section 12.4.1). But this is far from the end of the story. Levenberg– Marquardt assumes that it is possible to calculate both the first and second derivatives of the function to be minimized, that the starting parameter values are near the desired extremum of the cost function, and that the function is reasonably smooth. In practice these assumptions often do not apply, and so in this chapter we will look at ways to relax them. This chapter will cover the unconstrained optimization of a cost function; Chapter 17 will add restrictions to the search space for constrained optimization. Perhaps because nature must solve these kinds of optimization problems so frequently, algorithms in this chapter can have a kind of natural familiarity missing in most numerical methods. We'll see blobs (for the downhill simplex search), mass (momentum for avoiding local minima), temperature (via simulated annealing), and reproduction (with genetic algorithms). The cost of this kind of intuition is rigor. While there is some supporting theory, their real justification lies in their empirical performance. Consequently it's important to view them not as fixed received wisdom, but rather as a framework to guide further exploration.","PeriodicalId":430009,"journal":{"name":"Pricing and Revenue Optimization","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126986027","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"2 Introduction to Pricing and Revenue Optimization","authors":"","doi":"10.1515/9781503614260-003","DOIUrl":"https://doi.org/10.1515/9781503614260-003","url":null,"abstract":"","PeriodicalId":430009,"journal":{"name":"Pricing and Revenue Optimization","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121981508","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
求助内容：
应助结果提醒方式：