I took an early interest in computer viruses (see article in Abacus, 1987 and in 1989 co-invented a scheme called liveware which enables the distribution of databases, and the maintenance of distributed databases, using the transmission technique of computer viruses (reported in International journal. I have also worked on adaptively formed models for music prediction ( Interface: journal of New Music Research, 1990). Research support,250 pounds A programmable speech output peripheral src (uk science research council) 1979 8,600 pounds Synthesis of suprasegmental features using Votrax src 1979 3,200 pounds Interactive computer system using the Chinese language monotype International,200 pounds lpc vocoder implementation uk government Communications headquarters 1980 37,300. publications in Abacus acm operating Systems review acm sigoa newsletter ai and Society behaviour and Information Technology byte canadian Artificial Intelligence communications of the acm computer Bulletin Computer Graphics Computer journal Computer Music journal Computer Science Education Computers and Security computing Surveys Digital Processes Electronics.
Professional, resume, writers Cambridge kitchener, waterloo
In 1982/83, Dr Cleary and I discovered some fundamental results concerning adaptive modeling for data compression (published in ieee transactions on Information Theory, 1984 and invented a text compression method which outperforms virtually all known techniques by a large margin(emit typically takes only.2 bits/character. We have also published an influential study of bill arithmetic coding in Communications of the acm (1987 a survey of modeling for text compression ( Computing Surveys, 1989 and a major research book on adaptive text compression ( Text Compression, Prentice hall, 1990). Much of this work is with Dr Bell of the University of Canterbury, new zealand, a former postdoctoral fellow at Calgary, with whom I have worked on models for natural language text ( International journal of Man-Machine Studies, 1990) and the zero-frequency problem ( ieee. Compressed information retrieval systems and digital libraries Our work in text compression has book led to advances in compressed full-text retrieval systems. We developed new techniques for compressing a concordance (reported in the journal of Information Science, 1991 which is an inverted index of all words in a text, and it transpired that a compressed concordance gives substantial advances over the method of signature files, which represent. Along with research on the compression of document images ( Proceedings of the ieee, 1994 this work has culminated in a major book, managing Gigabytes: compressing and indexing documents and images. This is a substantial (430.) research text, along with a public-domain software system, which is already having a major impact on the area of information retrieval systems. Current directions are in investigating the design and construction of digital libraries, with papers presented at several international Digital Libraries conferences. Other research areas These include a (serious, src-funded) project on positional play in chess ( International journal of Man-Machine Studies, 1975, 1976 which has excited considerable interest; Chinese-language computer systems, funded by monotype International (1979 semiotics and computers; and text processing. An interesting accomplishment was the development of a hand-controlled speech toy (published in Wireless World, 1978, 1979) which attracted a large number of requests for circuit cards by would-be constructors and has been used by a developmental psychologist as a stimulus for autistic children. During I was heavily involved in Jade, a large group project at Calgary which aimed to construct an environment for the development of distributed software ( acm operating Systems review, 1983).
We have extended the idea to repetitive text editing, and have developed a theoretical approach to evaluating programs formed by example. Machine learning A number of technical advances in machine learning have been made, including a new algorithm for similarity-based learning (reported in the 1988 Machine learning conference a simplified presentation of explanation-based learning that removes it from its rather specialized niche and places. Contributions have also been made towards the rapprochement of knowledge acquisition and machine learning (reported in International journal of Man-Machine Studies, 1988, and ieee transactions on Systems, man and Cybernetics, 1989). Documentation systems In 1980 my book communicating with Microcomputers was published by Academic Press from camera-ready copy. During the production process I became intrigued with the manifest problems of existing document preparation systems and have developed a strong interest not only in batch document formatting languages and typography (papers in Software(emPractice and Experience, 1982, and International journal of Man-Machine Studies, 1985 but. This interest has extended to interactive systems which select documentation or advice appropriate to each individual user's needs, write based on long-term observations of their behavior, in the manner of an automated coach ( International journal of Man-Machine Studies, 1985). Text compression Data compression is modeling and prediction at work (see article in Abacus, 1986).
Out of this developed a general interest in user interface interaction styles ( Oxford Surveys in Information Technology, 1985). A notable adaptive interface is the reactive keyboard a device that accelerates typewritten communication with a computer reviews system by predicting what the user is going to type next (reported in Interacting with Computers, 1991). Macintosh (reported in ieee computer, 1990 pc, and Unix versions have been distributed worldwide as a computer interface for disabled people, and, according to users, have dramatically altered the way they use computers. A monograph on the system ( The reactive keyboard, c 1992) has been published by cambridge University Press. Programming by example Programming by example is the inference of programs from examples of their execution. We have created a novel system called Metamouse a user interface agent that learns to perform editing tasks within a drawing program (reported in Computer Graphics, 1989). The user specifies a procedure by supplying an example execution trace, manipulating objects directly on the screen and creating graphical tools where necessary to help make constraints explicit. Metamouse induces the procedure, identifying key features of individual program steps and disregarding coincidental events, and uses it to predict upcoming actions and perform them for the user, thereby reducing the tedium of repetitive graphical editing. The scheme has attracted a good deal of interest, including, for example, unsolicited donations from Apple computer to further the research.
Digital signal processing Work on speech synthesis naturally brought me into contact with speech analysis in the form of digital processing of signals. As well as developing special-purpose architectures for speech analysis, for which I was awarded a grant from the government Communications headquarters (1979-1981 i have worked on novel algorithms for signal processing, with a major paper in ieee transactions on Circuits and Systems (1981 and. Linear and non-linear identification techniques my separate interests in adaptive systems and speech converge in the area of identification of time series. I have applied linear techniques to unusual problems such as the identification of curves used in handwriting. As well as contributing to the development and analysis of algorithms for non-linear behavior sequence identification ( International journal of General Systems, 1979, 1980 i have used the techniques in human-computer interfaces which predict user entries (see below). A survey of methods for modeling behavior sequences appears in Future computer Systems (1987). Adaptive user interfaces Out of work on adaptive systems and behavior sequence identification sprang a strong interest in the potential of adaptation to simplify the user interface. A series of theoretical and experimental investigations demonstrated that (contrary to popular belief) interfaces can be designed to adapt automatically and successfully to the needs of individual users (reported in International journal of Man-Machine Studies, 1984, and Behaviour and Information Technology, 1985).
Waterloo, jobs - kitchener, waterloo, jobs
It encompasses the so-called two-armed bandit problem, of which I published a major survey ( journal of the Franklin Institute, 1976 resulting in an invitation by Professor cover, the world's leading authority in the area, to give a seminar at Stanford University (1977). The two-armed bandit problem epitomizes the conflict between system identification and control, where the cost of exploratory activity must be balanced against the cost of poor system knowledge. In bald form, it concerns an experimenter who is given two biased coins, whose biases are unknown, and must make tosses of either coin with the promise of getting 1 for heads but nothing for tails. His dilemma is that on the basis of a finite sample he can guess which coin is favored, but he may be wrong, and to increase confidence in his guess he must choose the apparently inferior coin sometimes. Constraints of finite time, finite memory (in the sense of remembering the results of the last few tosses only and finite state make the problem an interesting system-theoretic study, with potential application to any man-machine system which demands simultaneous exploration and control. It also includes adaptive control systems, the study of which led to an important theoretical paper in Information and Control (1977 an invited paper in the journal of Cybernetics and Systems Science (1977 the award of an Erskine fellowship to study in New zealand (1977.
Speech synthesis my contributions to this field have been diverse. I have worked on the design of speech synthesizers, for which I was consultant to the Essex Electronics Centre (1973-1974) and attracted an src research grant (1975-1977). My major contribution has been in the area of synthesis of prosodic features, which led to a key paper in Language and Speech (1977) and attracted another src grant (1979). Much of this work has proceeded in collaboration with the Phonetics Department at the University of Edinburgh, the department of Computer Science at the University of Calgary, bell-Northern Research at Montreal, bill the polish Academy of Sciences in poznan, and the department of Experimental Psychology. A major research book, principles of Computer Speech (Academic Press, 1982 contains the results of my own studies, as well as those of others. One of my papers was selected for inclusion in an ieee computer Society collection entitled End user facilities in the 1980's (1982).
D Elec Eng Dept, University of Canterbury, nz 1986 Robert Biddle. D computer Science dept, University of Canterbury, nz 1987 Tim Bell. D computer Science dept, University of Canterbury, nz 1988 John koegel. D computer Science dept, technical Univ of Graz, austria 1990 Norma fuller. D computer Science dept, University of Regina, canada 1991 pui-wing Wong. D systems Eng Dept, University of Waterloo, canada 1992 Alison lee.
D computer Science dept, University of Toronto, canada 1992 Phil Ngan. D computer Science dept, massey university, new zealand 1993 Chris Phillips. D computer Science dept, massey university, new zealand 1994 david Andreae. D computer Science dept, victoria university, new zealand 1993 paul Anderson. D computer Science dept, massey university, new zealand Membership in professional societies and editorial boards Fellow Association for Computing Machinery (ACM) Member American Society for Information Science (asis) Institute of Electrical Engineers (IEE) Institution of Electrical and Electronic Engineers (ieee) New zealand Computer Society (nzcs). In pursuit of this theme i have been drawn into adaptive text compression (using information about past text to encode upcoming characters machine learning (seeking ways to summarize, re-structure, and generalize past experience and user modeling (characterizing user behavior). The work represents the continuation of a substantial research effort which has already borne fruit in areas as diverse as computer speech, user modeling, adaptive interfaces, documentation browsing systems, data compression, machine learning, and knowledge acquisition. Theory of adaptive systems This is the subject of.
Archives - adventis Solutions
Department of Electrical Engineering Science, university of Essex, England, lecturer. Department of Electrical Engineering Science University of Essex, England Graduate Assistant Department of Mathematics, Statistics, and Computer Science, university of Calgary, canada visiting appointments and scholarships 1989, jun-Jul serc visiting Fellow University of Stirling, Scotland 1986, jan-Apr Erskine fellow University of Canterbury, new zealand 1985. D word identification in speech by phonetic analysis 1982 Rod Cuff. D database query using menus and natural language fragments 1984 saul Greenberg User modeling in interactive computer systems 1985 mike bonham viewing and formatting documents on-line 1985 Adrian Zissos generating advice by monitoring user behaviour 1985 roy masrani online conceptual analysis in Prolog 1988 John Darragh. D tool use, re-use, and organization in command-driven interfaces 1989 Thong Phan The equal-value search: accelerating search in function induction 1989 Dan mo learning text editing procedures from examples 1990 Darrell Conklin Prediction and entropy of music 1990 Antonija mitrovic Interactive induction of procedures (in. D Instructible agents 1994 Thong Phan. D function induction 1995 Brent Martin Instance-based learning: nearest neighbour with generalisation current Jamie littin mcms learning relational ripple-down remote rules current Matt Humphrey il A graphical notation for the design of information visualisations current Craig nevill-Manning il Detecting sequential structure current Stuart Inglis il current. D Elec Eng Dept, University of Canterbury, nz 1984 Bruce macDonald.
(unearned university of, cambridge, england 1970. University of, calgary, canada (Mathematics, Statistics, and Computer Science) 1969. (First Class, honours university of nature Cambridge, england (Mathematics) 1976, chartered, engineer. Institute of Electrical Engineers, london, employment 1991-present, professor. Department of Computer Science, university of waikato, new zealand, professor. Department of Computer Science, university of Calgary, canada. Head, department of, computer Science, university of Calgary, canada, senior Lecturer.
in Sign Up, schools. Cancel Please enter a username. Please enter a password. Citizenship, british, canadian, and New zealand, family Status. Married, two children, qualifications 1976,. University of, essex, England (Electrical Engineering Science) 1973.
Prior to that, i was responsible for driving and improving AdWords Budget Optimizer. Even earlier, i wrote probably the first method for large scale distributive blending of language models used for speech recognition and machine translation, and I was a technical lead of visual product search app for mobile phones. Launches include new products and features in Display ads, Image search, voice search, and Android. My core technical work is on par with other half of work i am doing: finding gaps in thinking, creating rough ideas, driving them and seeing them through. Edit, personal Details, other Works: Played Polly in "Chocky" based on the novel. John Wyndham - bbc radio 4 (1998). See more publicity listings: 2 Interviews, see more alternate names: Holly Grainger holliday granger holly Granger. Height: 5' 1" (1.55 m edit, did you know? Personal": guaranteed I'd love to do more independent films.
Available restaurant positions
Initiated, led, executed and productionized various efforts in some of the major focus areas for google. Currently, i am dual-role technical lead/manager in live, responsible for discovery and search of live streams on all/Google platforms. Prior to that, i was founding engineer in doubleClick database bid Manager (DBM) in New York office. Dbm (bidder and by side platform) connects world largest trading desks, agencies and advertisers with consolidated supply from 50 exchanges, processing millions of transactions in a second with 30ms latency. I helped grew video and mobile businesses from zero to 1B figure in under 30 months, and managed 18 engineers. Previously, i was an architect and implementer of a generic nearest neighbor search engine that powers similarity search in several google properties with tens of billions of data points. This engine includes an array of state of the art hashing and partitioning algorithms and high performance architecture and serving stack for approximate nearest neighbor search (k-nn).