We investigate the information complexity of mixed-integer convex optimization under different types of oracles. We establish new lower bounds for the standard first-order oracle, improving upon the previous best know...
详细信息
We investigate the information complexity of mixed-integer convex optimization under different types of oracles. We establish new lower bounds for the standard first-order oracle, improving upon the previous best known lower bound. This leaves only a lower order linear term (in the dimension) as the gap between the lower and upper bounds. This is derived as a corollary of a more fundamental "transfer" result that shows how lower bounds on information complexity of continuous convex optimization under different oracles can be transferred to the mixed-integer setting in a black-box manner. Further, we (to the best of our knowledge) initiate the study of, and obtain the first set of results on, information complexity under oracles that only reveal partial first-order information, e.g., where one can only make a binary query over the function value or subgradient at a given point. We give algorithms for (mixed-integer) convex optimization that work under these less informative oracles. We also give lower bounds showing that, for some of these oracles, every algorithm requires more iterations to achieve a target error compared to when complete first-order information is available. That is, these oracles are provably less informative than full first-order oracles for the purpose of optimization.
PurposeData transformation has prompted enterprises to rethink their strategic development. Scholars have frequently acknowledged the vast potential value of supply chain data and realised that simply owning data reso...
详细信息
PurposeData transformation has prompted enterprises to rethink their strategic development. Scholars have frequently acknowledged the vast potential value of supply chain data and realised that simply owning data resources cannot guarantee excellent innovation performance (IP). Therefore, this study focussed on the mediating and moderating issues between data-driven supply chain orientation (DDSCO) and IP. More specifically, the purpose was to explore (1) whether DDSCO promotes enterprise innovation through dynamic and improvisational capabilities and (2) how information complexity (INC) plays a moderating role between capabilities and ***/methodology/approachAn empirical study was performed using the results of a questionnaire survey, and a literature review was used to build the premises of this study. A sample was conducted on 296 Chinese enterprises, and the data collected were used to test the hypothesis by successive *** research has implications for the theoretical development of DDSCO, as well as the dynamic capabilities (DC) and improvisation capabilities (IC) in innovation strategic literature. The empirical results show that DDSCO has a direct, positive impact on both DC and IC, which thus positively impact IP. Meanwhile, IC has a negative moderating effect on the path joining DC and IP. Conversely, IC has a positive moderating effect on the path joining IC and *** limitations/implicationsAlthough this study has limitations, it also creates opportunities for future research. The survey comes from different industries, so the possibility of unique influences within industries cannot be ruled out. Second, the authors' survey is based on cross-sectional data, which allow for more comprehensive data verification in the future. Third, this study also provides opportunities for future research, because it proves that DC and IC, as partial mediators of DDSCO and IP, can mine other paths of the data-driven supply chain i
In a recent breakthrough paper Braverman et al. (in: STOC'13, pp 151-160, 2013) developed a local characterization for the zero-error information complexity in the two-party model, and used it to compute the exact...
详细信息
In a recent breakthrough paper Braverman et al. (in: STOC'13, pp 151-160, 2013) developed a local characterization for the zero-error information complexity in the two-party model, and used it to compute the exact internal and external information complexity of the 2-bit AND function. In this article, we extend their results on AND function to the multi-party number-in-hand model by proving that the generalization of their protocol has optimal internal and external information cost for certain natural distributions. Our proof has new components, and in particular, it fixes a minor gap in the proof of Braverman et al.
This paper studies the question of lower bounds on the number of neurons and examples necessary to program a given task into feedforward neural networks. We introduce the notion of information complexity of a network ...
详细信息
This paper studies the question of lower bounds on the number of neurons and examples necessary to program a given task into feedforward neural networks. We introduce the notion of information complexity of a network to complement that of neural complexity. Neural complexity deals with lower bounds for neural resources (numbers of neurons) needed by a network to perform a given task within a given tolerance. information complexity measures lower bounds for the information (i.e. number of examples) needed about the desired input-output function. We study the interaction of the two complexities, and so lower bounds for the complexity of building and then programming feed-forward nets for given tasks. We show something unexpected a priori-the interaction of the two can be simply bounded, so that they can be studied essentially independently. We construct radial basis function (RBF) algorithms of order n(3) that are information-optimal, and give example applications. (C) 2000 Elsevier Science Ltd. all rights reserved.
Two parties observing correlated random variables seek to run an interactive communication protocol. How many bits must they exchange to simulate the protocol, namely to produce a view with a joint distribution within...
详细信息
Two parties observing correlated random variables seek to run an interactive communication protocol. How many bits must they exchange to simulate the protocol, namely to produce a view with a joint distribution within a fixed statistical distance of the joint distribution of the input and the transcript of the original protocol? We present an information spectrum approach for this problem whereby the information complexity of the protocol is replaced by its information complexity density. Our single-shot bounds relate the communication complexity of simulating a protocol to tail hounds for information complexity density. As a consequence, we obtain a strong converse and characterize the second-order asymptotic term in communication complexity for independent and identically distributed observation sequences. Furthermore, we obtain a general formula for the rate of communication complexity, which applies to any sequence of observations and protocols. Connections with results from theoretical computer science and implications for the function computation problem are discussed.
This paper is a lecture note accompanying the 19th Takagi Lectures lectures in July 2017 at Kyoto *** give a high-level overview of information complexity theory and its connections to communication *** then discuss s...
详细信息
This paper is a lecture note accompanying the 19th Takagi Lectures lectures in July 2017 at Kyoto *** give a high-level overview of information complexity theory and its connections to communication *** then discuss some fundamental properties of information complexity, and applications to direct sum theorems and to exact communication bounds. We conclude with some open questions and directions.
We investigate the information complexity of mixed-integer convex optimization under different kinds of oracles. We establish new lower bounds for the standard first-order oracle, improving upon the previous best know...
详细信息
ISBN:
(数字)9783031327261
ISBN:
(纸本)9783031327254;9783031327261
We investigate the information complexity of mixed-integer convex optimization under different kinds of oracles. We establish new lower bounds for the standard first-order oracle, improving upon the previous best known lower bound. This leaves only a lower order linear term (in the dimension) as the gap between the lower and upper bounds. Further, we prove the first set of results in the literature (to the best of our knowledge) on information complexity with respect to oracles based on first-order information but restricted to binary queries, and discuss various special cases of interest thereof.
In this paper, we investigate information complexity of integration of the Sobolev classes with bounded mixed derivative in the Monte Carlo setting. We develop Monte Carlo algorithms for integration of functions from ...
详细信息
ISBN:
(纸本)9783037852842
In this paper, we investigate information complexity of integration of the Sobolev classes with bounded mixed derivative in the Monte Carlo setting. We develop Monte Carlo algorithms for integration of functions from these classes and analyze their convergence rates. Comparing our result with the known convergence rates in the deterministic setting, we see that Monte Carlo algorithms bring a essential speed-up.
In a recent breakthrough paper Braverman et al. (in: STOC'13, pp 151-160, 2013) developed a local characterization for the zero-error information complexity in the two-party model, and used it to compute the exact...
详细信息
ISBN:
(纸本)9783319623887
In a recent breakthrough paper Braverman et al. (in: STOC'13, pp 151-160, 2013) developed a local characterization for the zero-error information complexity in the two-party model, and used it to compute the exact internal and external information complexity of the 2-bit AND function. In this article, we extend their results on AND function to the multi-party number-in-hand model by proving that the generalization of their protocol has optimal internal and external information cost for certain natural distributions. Our proof has new components, and in particular, it fixes a minor gap in the proof of Braverman et al.
作者:
Xing, JingFAA
Civil Aerosp Med Inst Oklahoma City OK 73125 USA
Air traffic controllers use visual displays to interact with various automation systems. information complexity in those systems may cause controllers to miss or misinterpret visual data, thereby reducing safety. The ...
详细信息
ISBN:
(纸本)9783540731092
Air traffic controllers use visual displays to interact with various automation systems. information complexity in those systems may cause controllers to miss or misinterpret visual data, thereby reducing safety. The purpose of this study was to answer three basic questions: 1) What constitutes information complexity in automation displays? 2) How complex is "too complex" for controllers? 3) Can we objectively measure information complexity in the displays? We first developed a general framework for measuring information complexity. The framework reduces the concept of complexity into three underlying factors: quantity, variety, and the relations between basic information elements;each factor is evaluated at three generic stages of human information processing: perception, cognition, and action. We then developed nine metrics of display complexity, each measuring the effects of a complexity factor on information processing at a given stage. These metrics provide an objective method to evaluate automation displays for acquisition and design prototypes.
暂无评论