ART-EMAP: A Neural Network Architecture for Learning and Prediction by Evidence Accumulation

This paper introduces ART-EMAP, a neural architecture that uses spatial and temporal evidence accumulation to extend the capabilities of fuzzy ARTMAP. ART-EMAP combines supervised and unsupervised learning and a medium-term memory process to accomplish stable pattern category recognition in a noisy input environment. The ART-EMAP system features (i) distributed pattern registration at a view category field; (ii) a decision criterion for mapping between view and object categories which can delay categorization of ambiguous objects and trigger an evidence accumulation process when faced with a low confidence prediction; (iii) a process that accumulates evidence at a medium-term memory (MTM) field; and (iv) an unsupervised learning algorithm to fine-tune performance after a limited initial period of supervised network training. ART-EMAP dynamics are illustrated with a benchmark simulation example. Applications include 3-D object recognition from a series of ambiguous 2-D views.

ART-EMAP: A Neural Network Architecture for Object Recognition by Evidence Accumulation

A new neural network architecture is introduced for the recognition of pattern classes after supervised and unsupervised learning. Applications include spatio-temporal image understanding and prediction and 3-D object recognition from a series of ambiguous 2-D views. The architecture, called ART-EMAP, achieves a synthesis of adaptive resonance theory (ART) and spatial and temporal evidence integration for dynamic predictive mapping (EMAP). ART-EMAP extends the capabilities of fuzzy ARTMAP in four incremental stages. Stage 1 introduces distributed pattern representation at a view category field. Stage 2 adds a decision criterion to the mapping between view and object categories, delaying identification of ambiguous objects when faced with a low confidence prediction. Stage 3 augments the system with a field where evidence accumulates in medium-term memory (MTM). Stage 4 adds an unsupervised learning process to fine-tune performance after the limited initial period of supervised network training. Each ART-EMAP stage is illustrated with a benchmark simulation example, using both noisy and noise-free data. A concluding set of simulations demonstrate ART-EMAP performance on a difficult 3-D object recognition problem.

3-D Object Recognition by the ART-EMAP Evidence Accumulation Network

ART-EMAP synthesizes adaptive resonance theory (AHT) and spatial and temporal evidence integration for dynamic predictive mapping (EMAP). The network extends the capabilities of fuzzy ARTMAP in four incremental stages. Stage I introduces distributed pattern representation at a view category field. Stage 2 adds a decision criterion to the mapping between view and object categories, delaying identification of ambiguous objects when faced with a low confidence prediction. Stage 3 augments the system with a field where evidence accumulates in medium-term memory (MTM). Stage 4 adds an unsupervised learning process to fine-tune performance after the limited initial period of supervised network training. Simulations of the four ART-EMAP stages demonstrate performance on a difficult 3-D object recognition problem.

Archbishop William King (1650-1729) and the Constitution of Church and State

Accepted Version

A SURVEY OF FRENCH KEYBOARD MUSIC FROM 1880 TO 1950

French music flourished from the last quarter of the nineteenth century into the early twentieth century, especially in the genres of opera and orchestral music. Although French keyboard music enjoyed less popularity, being overshadowed by these predominant genres, prominent impressionist figures Claude Debussy (1862-1918) and Maurice Ravel (1875-1937) . . brought its revival to the French music. Scholars consider Debussy to be a frontrunner of Impressionism, and his influence had a major impact on subsequent composers. As a result of his popularity, other significant works by French composers seem to be overlooked by pianists and audiences and are not as often performed. Because keyboard works by Debussy and Ravel are a popular performance choice among pianists, I was eager to examine music by other French composers. Through my resea,rch, I found many great works that warranted further study and deserve a place in the keyboard repertoire. This recording project contains works by lesser-known French composers written between the years of 1880 and 1950, namely Emmanuel Chabrier (1841-1894), Gabriel-Urbain Faure (1845-1924), Charles Koechlin (1867-1950), Albert Roussel (1869-1937), Erik Satie (1866-1925), Francis Poulenc (1899-1963), Darius Milhaud (1892-1974), Robert Casadesus (1899-1972) and Henri Dutilleux (b.1916). Since piano repertoire is abundant, it is sometimes difficult to create a performance program. Therefore, it frequently becomes the default to choosing familiar repertoire rather than using the opportunity to expand the repertoire. As a pianist, I feel responsible to search for hidden musical treasures with which pianists and audiences alike are not so well acquainted. This recording project explores nine lesser-known French compositions written between 1880 and 1950. I expect this to be an opportunity to introduce both pianists and audiences to outstanding but unfamiliar works by French composers. This dissertation was recorded on two compact discs in Dekelboum Concert Hall at Clarice Smith Performing Arts Center of the University of Maryland. The recordings are archived in the University of Maryland Library.

[Recording review] William Herschel, Symphonies Nos. 2, 8, 12, 13, 14, 17

Tony Mann provides a review of the recording: William Herschel, Symphonies Nos. 2, 8, 12, 13, 14, 17, soloists, London Mozart Players / Matthias Bamert, Chandos 10048, 2003, (68 mins). [CD]

Perspectives and Integration in SOLAS Science

Why a chapter on Perspectives and Integration in SOLAS Science in this book? SOLAS science by its nature deals with interactions that occur: across a wide spectrum of time and space scales, involve gases and particles, between the ocean and the atmosphere, across many disciplines including chemistry, biology, optics, physics, mathematics, computing, socio-economics and consequently interactions between many different scientists and across scientific generations. This chapter provides a guide through the remarkable diversity of cross-cutting approaches and tools in the gigantic puzzle of the SOLAS realm. Here we overview the existing prime components of atmospheric and oceanic observing systems, with the acquisition of ocean–atmosphere observables either from in situ or from satellites, the rich hierarchy of models to test our knowledge of Earth System functioning, and the tremendous efforts accomplished over the last decade within the COST Action 735 and SOLAS Integration project frameworks to understand, as best we can, the current physical and biogeochemical state of the atmosphere and ocean commons. A few SOLAS integrative studies illustrate the full meaning of interactions, paving the way for even tighter connections between thematic fields. Ultimately, SOLAS research will also develop with an enhanced consideration of societal demand while preserving fundamental research coherency.