Effects of the user model on simulation-based learning of dialogue strategies

Over the past decade, a variety of user models have been proposed for user simulation-based reinforcement-learning of dialogue strategies. However, the strategies learned with these models are rarely evaluated in actual user trials and it remains unclear how the choice of user model affects the quality of the learned strategy. In particular, the degree to which strategies learned with a user model generalise to real user populations has not be investigated. This paper presents a series of experiments that qualitatively and quantitatively examine the effect of the user model on the learned strategy. Our results show that the performance and characteristics of the strategy are in fact highly dependent on the user model. Furthermore, a policy trained with a poor user model may appear to perform well when tested with the same model, but fail when tested with a more sophisticated user model. This raises significant doubts about the current practice of learning and evaluating strategies with the same user model. The paper further investigates a new technique for testing and comparing strategies directly on real human-machine dialogues, thereby avoiding any evaluation bias introduced by the user model. © 2005 IEEE.

Model based learning of sigma points in unscented Kalman filtering

The unscented Kalman filter (UKF) is a widely used method in control and time series applications. The UKF suffers from arbitrary parameters necessary for a step known as sigma point placement, causing it to perform poorly in nonlinear problems. We show how to treat sigma point placement in a UKF as a learning problem in a model based view. We demonstrate that learning to place the sigma points correctly from data can make sigma point collapse much less likely. Learning can result in a significant increase in predictive performance over default settings of the parameters in the UKF and other filters designed to avoid the problems of the UKF, such as the GP-ADF. At the same time, we maintain a lower computational complexity than the other methods. We call our method UKF-L. ©2010 IEEE.

HACCP based management system for improving quality of freshwater giant prawn, Macrobrachium rosenbergii

The study was conducted on the present status of HACCP based quality management system of golda, Macrobrachium rosenbergii farms in Fulpur region of Mymensingh. Information was collected on general condition of farms, culture systems and post-harvest quality management. In almost all farms, there is no or inadequate infrastructure facilities such as, road access, electric supply, telecommunications, ice, feed storage facility, vehicle for golda transportation, washing and toilet facilities. The problems associated with sanitation and hygiene was: widespread use of cow dung, poultry manure and construction of open toilet within the vicinity of prawn culture pond. Different grades of commercially available and locally prepared feeds were used for golda culture in the pond. Golda post-larvae (PL) of 40-50 days old were stocked with carp species. The price of golda PL ranged from Tk. 1.00 to Tk. 1.25/piece. The pond size varied from 50 decimal (0.2 ha) to 2.5 acre (1.0 ha) with an average depth of 2-2.5 m. The culture period of golda varied from April-May to November-December and survival rate ranged between 75 and 80%. Production of golda varied from 250-500 kg/acre (625-1,250 kg/ha). Harvested golda were transported to city market within 4 h. Two size grading were generally followed during pricing, e.g. Tk. 500 to 550/kg for >100 g size and Tk. 300/kg for <100 g size. The cost-benefit ratio was found to remain around 1:1.25 depending on availability of PL. Water quality parameters such as, water temperature, pH, dissolved oxygen, total alkalinity and chlorophyll a in five golda farms in Fulpur region were monitored. Water temperature ranged from 29°C to 33°C, dissolved oxygen from 2.28 to 4.13 mg/l, pH between 6.65 and 7.94, alkalinity from 44 to 70 mg/l and chlorophyll a concentration from 61.88 to 102.34 µg/l in the five investigated ponds. The Aerobic Plate Count (APC) of the water sample was within the range of 2.0x10^6 - 2.96x10^7 CFU/ml and of soil samples within the range of 6.9x10^6 - 7.73x10^6 CFU/g. Streptococcus sp., Bacillus sp., Escherichia coli, Staphylococcus sp., Pseudomonas sp. and Salmonella sp. were isolated from pond water and sediment. Different feed samples used for golda was analyzed for proximate composition. Moisture content ranged around 14.14-21.22%, crude protein 20.55-44.1%, lipid 4.67-12.54% and ash 9.7-27.69%. The TVB-N values and peroxide values of feeds used as starter, grower and fish meal were found within the acceptable ranges and samples were free from pathogenic organisms. A training was organized for the golda farmers on HACCP, water quality and post-harvest quality management of prawn.

Computer-based selection of joining processes

The assembly of any manufactured product involves joining. This paper describes ways of selecting processes for joining. The method allows discrimination of the joint geometry, joint loading, material, and other attributes of the joint itself, identifying the subset of available processes capable of meeting a given set of design constraints. A relational database containing data-tables for joining processes, materials to be joined, and joint geometry and mode of loading, allows the attributes of each of these to be stored in an appropriate format, and permits links to be created between those that are related. A search engine isolates the processes that meet design requirements on material, joint geometry and loading. The method is illustrated in Part 2 by case studies, utilising software that embodies the method.

Proceedings of an international workshop on the value of computer-based engineering tools, held in Cambridge in April 1997

There is increasing adoption of computer-based tools to support the product development process. Tolls include computer-aided design, computer-aided manufacture, systems engineering and product data management systems. The fact that companies choose to invest in tools might be regarded as evidence that tools, in aggregate, are perceived to possess business value through their application to engineering activities. Yet the ways in which value accrues from tool technology are poorly understood.

This report records the proceedings of an international workshop during which some novel approaches to improving our understanding of this problem of tool valuation were presented and debated. The value of methods and processes were also discussed. The workshop brought together British, Dutch, German and Italian researchers. The presenters included speakers from industry and academia (the University of Cambridge, the University of Magdeburg and the Politechnico de Torino)

The work presented showed great variety. Research methods include case studies, questionnaires, statistical analysis, semi-structured interviews, deduction, inductive reasoning, the recording of anecdotes and analogies. The presentations drew on financial investment theory, the industrial experience of workshop participants, discussions with students developing tools, modern economic theories and speculation on the effects of company capabilities.