Learning market prices in real-time supply chain management

This paper proposes a model for dynamic pricing that combines knowledge of production capacity and existing commitments, reasoning about uncertainty and learning of market conditions in an attempt to optimise expected profits. In particular, the market conditions are represented as a set of probabilities over the success rate of product prices, and those prices are learned online as the market develops. The dynamic pricing model is integrated into a real-time supply chain management agent using the Trading Agent Competition Supply Chain Management game as a test framework. We evaluate the agent experimentally in competition with other supply chain agents, and demonstrate the benefits of incorporating more market data into the dynamic pricing mechanism.     

An empirical investigation of anti-spyware software adoption: A multitheoretical perspective

We performed an empirical investigation of factors affecting an individual's decision to adopt anti-spyware software. Our results suggested that an individual's attitude, subjective norm, perceived behavioral control, and denial of responsibility significantly affected anti-spyware adoption intention. Also, relative advantage and compatibility showed a significant effect on attitude, visibility, and image on subjective norm, and trialability, self-efficacy, and computing capacity on perceived behavioral control. Interestingly, moral obligation, ease of use, and perceived cost were not as significant as was originally expected.     

Simultaneous measurement of temperature and strain with long period grating pairs using low resolution detection

An investigation showing the performance of compact long period grating (LPG) pairs under demanding experimental conditions, making comparisons with the use of single LPGs for the simultaneous measurement of strain and temperature is presented. In this work, a LPG pair (LPGP) sensor comprising two weak (2.5 dB each) LPGs, 13 mm apart, was compared against the performance of two single LPG sensors (LPG1 & LPG2) with coupling strengths of 2.5 dB and 12.5 dB, respectively. All the LPG sensors used were subjected to extreme conditions arising from the fabrication, by choosing a high amplitude mask period with short physical length LPGs, and the high temperature annealing processes. In addition, a low resolution detection instrument was used to capture the spectra under study. Results have shown that the LPGP is 40 times better than the first single LPG used (LPG1) and three times better than the second LPG (LPG2). Using the error analysis methodology of Brady et al., the temperature and strain errors due to the uncertainty of the wavelength measurement for each of the LPG1, LPG2, and LPGP were compared with the typical performance of various fibre Bragg grating (FBG)-based sensor schemes under similar testing conditions. It was found that LPGPs offer the best performance especially when using low resolution detection instrumentation.     

Companion Cognitive Systems: Design Goals and Lessons Learned So Far

The Companion cognitive architecture supports experiments in achieving human-level intelligence. This article describes seven key design goals of Companions, relating them to properties of human reasoning and learning, and to engineering concerns raised by attempting to build large-scale cognitive systems. We summarize our experiences with Companions in two domains: test taking and game playing.     

A Ramsey theorem in Boyer-Moore logic

We use the Boyer-Moore Prover, Nqthm, to verify the Paris-Harrington version of Ramsey's theorem. The proof we verify is a modification of the one given by Ketonen and Solovay. The theorem is not provable in Peano Arithmetic, and one key step in the proof requires ₀ induction.Author supported by NSF Grant DMS-9100665.     

Picbreeder: a case study in collaborative evolutionary exploration of design space

For domains in which fitness is subjective or difficult to express formally, interactive evolutionary computation (IEC) is a natural choice. It is possible that a collaborative process combining feedback from multiple users can improve the quality and quantity of generated artifacts. Picbreeder, a large-scale online experiment in collaborative interactive evolution (CIE), explores this potential. Picbreeder is an online community in which users can evolve and share images, and most importantly, continue evolving others' images. Through this process of branching from other images, and through continually increasing image complexity made possible by the underlying neuroevolution of augmenting topologies (NEAT) algorithm, evolved images proliferate unlike in any other current IEC system. This paper discusses not only the strengths of the Picbreeder approach, but its challenges and shortcomings as well, in the hope that lessons learned will inform the design of future CIE systems.     

Three heuristic procedures for the stochastic,two-machine flow shop problem

Although the deterministic flow shop model is one of the most widely studied problems in scheduling theory, its stochastic analog has remained a challenge. No computationally efficient optimization procedure exists even for the general two-machine version. In this paper, we describe three heuristic procedures for the stochastic, two-machine flow shop problem and report on computational experiments that compare their effectiveness. We focus on heuristic procedures that can be adapted for dispatching without the need for computer simulation or computer-based search. We find that all three procedures are capable of quickly generating solutions close to the best known sequences, which were obtained by extensive search.     

A scalable delay based analytical framework for CSMA/CA wireless mesh networks

We present an analytical framework for the performance analysis of CSMA/CA based wireless mesh networks. This framework can provide an accurate throughput-delay evaluation for both saturated and unsaturated cases. An efficient algorithm that determines the collision domain for each node based on both the interference range and routing in the network is presented. As another important application of this framework, we develop an analytic model that enables us to obtain closed form expressions for delay in terms of multipath routing variables. A flow-deviation algorithm is used to derive the optimal flow over a given set of routes for any number of classes. The model takes into account the effects of neighbor interference and hidden terminals, and tools are provided to make it feasible for the performance analysis and optimization of large-scale networks. Numerical results are presented for different network topologies and compared with simulation studies.