Matching Items (187)
Filtering by

Clear all filters

190780-Thumbnail Image.png
Description
Artificial Intelligence (AI) and Machine Learning (ML) techniques have come a long way since their inception and have been used to build intelligent systems for a wide range of applications in everyday life. However they are very computationintensive and require transfer of large volume of data from memory to the

Artificial Intelligence (AI) and Machine Learning (ML) techniques have come a long way since their inception and have been used to build intelligent systems for a wide range of applications in everyday life. However they are very computationintensive and require transfer of large volume of data from memory to the computation units. This memory access time constitute significant part of the computational latency and a performance bottleneck. To address this limitation and the ever-growing demand for implementation in hand-held and edge-devices, In-memory computing (IMC) based AI/ML hardware accelerators have emerged. First, the dissertation presents an IMC static random access memory (SRAM) based hardware modeling and optimization framework. A unified systematic study closely models the IMC hardware, and investigates how a number of design variables and non-idealities (e.g. device mismatch and ADC quantization) affect the Deep Neural Network (DNN) accuracy of the IMC design. The framework allows co-optimized selection of different design variables accounting for sources of noise in IMC hardware and robust implementation of a high accuracy DNN. Next, it presents a kNN hardware accelerator in 65nm Complementary Metal-Oxide-Semiconductor (CMOS) technology. The accelerator combines an IMC SRAM that is developed for binarized deep neural networks and other digital hardware that performs top-k sorting. The simulated k Nearest Neighbor accelerator design processes up to 17.9 million query vectors per second while consuming 11.8 mW, demonstrating >4.8× energy-efficiency improvement over prior works. This dissertation also presents a novel floating-point precision IMC (FP-IMC) macro with a hybrid architecture that configurably supports two Floating Point (FP) precisions. Implementing FP precision MAC has been a challenge owing to its complexity. The design is implemented on 28nm CMOS, and taped-out on chip demonstrating 12.1 TFLOPS/W and 66.1 TFLOPS/W for 8-bit Floating Point (FP8) and Block Floating point (BF8) respectively. Finally, another iteration of the FP design is presented that is modeled to support multiple precision modes from FP8 up to FP32. Two approaches to the architectural design were compared illustrating the throughput-area overhead trade-off. The simulated design shows a 2.1 × normalized energy-efficiency compared to the on-chip implementation of the FP-IMC.
ContributorsSaikia, Jyotishman (Author) / Seo, Jae-Sun (Thesis advisor) / Chakrabarti, Chaitali (Thesis advisor) / Fan, Deliang (Committee member) / Cao, Yu (Committee member) / Arizona State University (Publisher)
Created2023
190975-Thumbnail Image.png
Description
This thesis addresses the problems of (a) scheduling multiple streaming jobs with soft deadline constraints to minimize the risk/energy consumption in heterogeneous Systems-on-chip (SoCs), and (b) training a neural network model with high accuracy and low training time using split federated learning (SFL) with heterogeneous clients. Designing a scheduler for

This thesis addresses the problems of (a) scheduling multiple streaming jobs with soft deadline constraints to minimize the risk/energy consumption in heterogeneous Systems-on-chip (SoCs), and (b) training a neural network model with high accuracy and low training time using split federated learning (SFL) with heterogeneous clients. Designing a scheduler for heterogeneous SoC SoCs built with different types of processing elements (PEs) is quite challenging, especially when it has to balance the conflicting requirements of low energy consumption, low risk, and high throughput for randomly streaming jobs at run time. Two probabilistic deadline-aware schedulers are designed for heterogeneous SoCs for such jobs with soft deadline constraints with the goals of optimizing job-level risk and energy efficiency. The key idea of the probabilistic scheduler is to calculate the task-to-PE allocation probabilities when a job arrives in the system. This allocation probability, generated by manually designed or neural network (NN) based allocation function, is used to compute the intra-job and inter-job contentions to derive the task-level slack. The tasks are allocated to the PEs that can complete the task within the task-level slack with minimum risk or minimum energy consumption. SFL is an edge-friendly decentralized NN training scheme, where the model is split and only a small client-side model is trained in the clients. The communication overhead in SFL is significant since the intermediate activations and gradients of every sample are transmitted in every epoch. Two communication reduction methods have been proposed, namely, loss-aware selective updating to reduce the number of training epochs and bottleneck layer (BL) to reduce the feature size.Next, the SFL system is trained with heterogeneous clients having different data rates and operating on non-IID data. The communication time of clients in low-end group with slow data rates dominates the training time. To reduce the training time without sacrificing accuracy significantly, HeteroSFL is built with HetBL and bi- directional knowledge sharing (BDKS). HetBL compresses data with different factors in low- and high-end groups using narrow and wide bottleneck layers respectively. BDKS is proposed to mitigate the label distribution skew across different groups. BDKS can also be applied in Federated Learning to address the label distribution skew.
ContributorsChen, Xing (Author) / Chakrabarti, Chaitali (Thesis advisor, Committee member) / Ogras, Umit (Committee member) / Fan, Deliang (Committee member) / Zhang, Jeff (Committee member) / Arizona State University (Publisher)
Created2023
155502-Thumbnail Image.png
Description
The concert vocalise, a dazzling wordless vocal etude intended for performance, is largely a phenomenon of the twentieth century. Made famous by composers such as Sergei Rachmaninoff and Maurice Ravel, the concert vocalise is generally a short, non-programmatic work with a relatively simple form. In contrast, Nikolai Medtner’s

The concert vocalise, a dazzling wordless vocal etude intended for performance, is largely a phenomenon of the twentieth century. Made famous by composers such as Sergei Rachmaninoff and Maurice Ravel, the concert vocalise is generally a short, non-programmatic work with a relatively simple form. In contrast, Nikolai Medtner’s two monumental Op. 41 vocalises, the Sonata-Vocalise mit einem Motto “Geweihter Platz and the Suite Vocalise, are staggering in their length and formal complexity. They are also programmatically conceived, sharing the Goethe poem “Geweihter Platz” as their inspiration.

The innovation of adding a textual element to a traditionally textless genre introduces a tantalizing new layer of complexity that demands further research and exploration. However, as with any innovation, it also offers new challenges to performers wishing to program either or both works. Current scholarship has yet to offer any kind of in-depth analysis of either work, leaving questions as to the structural and motivic elements which bind these large works together, not to mention questions related to exactly how Medtner addresses the challenge of linking specific parts of Goethe’s text to the textless portions of music. Furthermore, neither work is considered standard repertoire, and recordings and performances are limited, leaving aspiring performers in something of an informational desert.

In this paper, I endeavor to fill this informational gap for performers and scholars alike by providing them with a brief biography of Medtner, an outline of the development of the concert vocalise genre, and the background of the Goethe poem that inspired Medtner. Then my in-depth analyses reveal underlying structural, motivic, and programmatic links both within and between the works. Finally, my performer’s guide, based on the analyses and my experience performing both works, offers suggestions regarding the interpretational, ensemble, and technical challenges presented by these great works.
ContributorsUhl, Nathan Leonard (Author) / Campbell, Andrew (Thesis advisor) / Carpenter, Ellon (Thesis advisor) / FitzPatrick, Carole (Committee member) / Ryan, Russell (Committee member) / Arizona State University (Publisher)
Created2017
158030-Thumbnail Image.png
Description
The pipe organ, a musical instrument historically, culturally, and conceptually associated with Christian worship, is being negatively affected in terms of condition and continued use in the United States by rising rates of secularity, declining mainline Protestant and Catholic adherents, declining worship attendance, and, most directly, the closure of church

The pipe organ, a musical instrument historically, culturally, and conceptually associated with Christian worship, is being negatively affected in terms of condition and continued use in the United States by rising rates of secularity, declining mainline Protestant and Catholic adherents, declining worship attendance, and, most directly, the closure of church buildings. No scholarly research addresses the relationship of secularization and the organ, and no professionals in the field acknowledge its seriousness or have presented plans to counter it.

This paper lays the groundwork for future research while exploring the landscape of the organ’s possible secular uses. The organ’s relationship to secularization is defined through the interdisciplinary lens of secular studies, bolstered through an exploration of its past. This thesis analyzes the use of the organ in secularized churches in the United States through case studies of fourteen organs in thirteen former churches. While these examples reveal some promising adaptive reuses of church buildings and their organs, the prevailing conclusion is that the instrument’s future is severely endangered. There are few paid secular positions and insignificant educational opportunities that stem from secularized churches. The public lacks exposure to the instrument because of the infrequency of organ-related events.

Yet because the organ’s principal aesthetic is not Christian but communal, the instrument has the potential to thrive in secular contexts. This reframing and often literal repositioning requires stronger leadership: organizations and individuals promoting the organ must be proactive in working with the new owners of secularized churches to help them incorporate the instrument in new, revitalized contexts. A dynamic future for the organ requires the creative work of many.
ContributorsMeszler, Alexander Francis (Author) / Marshall, Kimberly (Thesis advisor) / Ryan, Russell (Committee member) / Schmelz, Peter (Committee member) / Arizona State University (Publisher)
Created2020
158188-Thumbnail Image.png
Description
Lu Pei (b. 1956) is a celebrated Chinese American composer who currently serves as a composition professor at the Shanghai Conservatory of Music. His work is known for its varied use of Chinese folk tunes. He lived in the United States a long time, which cultivated within him abundant Western

Lu Pei (b. 1956) is a celebrated Chinese American composer who currently serves as a composition professor at the Shanghai Conservatory of Music. His work is known for its varied use of Chinese folk tunes. He lived in the United States a long time, which cultivated within him abundant Western musical influences. The sound of multi-ethnic elements is greatly reflected in his own music. Writing an opera has always been his ultimate dream as a composer, and after many years of work, the opera The Third of March was completed and premiered in 2018 in Guangxi. It has received wide acclaim and has been a favorite of younger listeners. Lu Pei aims to bring young people to modern music and the Chinese opera, not only using traditional Chinese musical elements, but also adopts “reinvented” modern Western musical styles, giving a new identity to the Chinese opera is the main foci of The Third of March.

To prepare for my performer’s guide to The Third of March, I will discuss Lu Pei’s inspirations from the Guangxi Song Fairs, and the music and culture of the Zhuang people surrounding the date in the Chinese lunar calendar, March Third. For Westerners unfamiliar with Lu Pei’s music, I will briefly introduce the compositional blending of Western and Chinese musical styles with a section about Chinese composers active in the United States, Chen Yi (b. 1953), and Tan Dun (b. 1957). I will also include a brief outline of the history of Chinese opera development, and Lu Pei’s compositional concepts and the background of the opera The Third of March will be discussed.

My performer’s guide, the primary focus of this project, will begin by stressing Lu Pei’s adoption of different Chinese folk songs and Western compositional elements. These techniques clearly gave the piece a unique stylistic identity. I will give a brief overview of the Chinese language diction in International Phonetic Alphabet. Finally, the qualities of the main arias in the opera, and some of the Chinese operatic techniques for singers, and their special effects, will be explored.
ContributorsBan, Banlingyu (Author) / FitzPatrick, Carole (Thesis advisor) / Norton, Kay (Committee member) / Ryan, Russell (Committee member) / Arizona State University (Publisher)
Created2020
158210-Thumbnail Image.png
Description
ABSTRACT

Emilie Mayer (1812-1883) was a prolific composer whose musical works, which encompassed eight symphonies, four overtures, an opera, dozens of sonatas, eight string quartets, solo piano works, and nearly 130 songs for solo voice or vocal quartet, were performed in the foremost concert halls in Berlin and across Germany. She

ABSTRACT

Emilie Mayer (1812-1883) was a prolific composer whose musical works, which encompassed eight symphonies, four overtures, an opera, dozens of sonatas, eight string quartets, solo piano works, and nearly 130 songs for solo voice or vocal quartet, were performed in the foremost concert halls in Berlin and across Germany. She studied with lauded teachers: Carl Loewe (1796-1869), Adolph Bernhard Marx (1795-1866), and Wilhelm Wieprecht (1802-1872). Her talent was applauded by audiences and critics wrote favorably, despite their reservations about women composers. However, even with this unusual pedigree, Mayer’s works nearly disappeared from concert stages after her death. How did this happen? This study aims to answer this question and will delve into Emilie Mayer’s life and works in context with the prejudices against female composers at the time, in order to determine how those biases have shaped the classical canon. Included is an in-depth stylistic analysis of Mayer’s surviving seven Lieder, along-side comparisons to similar works of other composers. In addition, appendices present Mayer’s remaining Lieder in a new, modernized edition, with selected songs transposed for better accessibility for lower voices. Relative lack of female representation in modern-day concert halls and music history books correlates to previous misconceptions of female composers. Studying the works of Emilie Mayer will support her addition to the classical repertoire, help correct the male-gendered canon that persists, and help modern female composers realize their history is not confined to a footnote.
ContributorsSadownik, Stephanie Beth (Author) / Weiss, Stephanie (Thesis advisor) / Norton, Kay (Thesis advisor) / Ryan, Russell (Committee member) / Arizona State University (Publisher)
Created2020
158427-Thumbnail Image.png
Description
ABSTRACT



"The Soul Unto Itself," a chamber music song cycle, was commissioned by the author, Rosa LoGiudice, and composed by William Clay, a doctoral candidate in composition at Arizona State University. The cycle was conceived and composed in the summer and fall

ABSTRACT



"The Soul Unto Itself," a chamber music song cycle, was commissioned by the author, Rosa LoGiudice, and composed by William Clay, a doctoral candidate in composition at Arizona State University. The cycle was conceived and composed in the summer and fall of 2019. The chamber ensemble was a sextet comprised of Megan Law, mezzo-soprano, Kristi Hanno, clarinet, Emilio Vazquez, violin, Rittika Gambhir, bassoon, Nathaniel De la Cruz, double bass, and Rosa LoGiudice, piano, all based in Tempe, Arizona. The song cycle was premiered in a lecture recital on December 8, 2019 at Hammer and Strings Conservatory in Gilbert, AZ.

"The Soul Unto Itself" is a cycle of six songs based on poems of Emily Dickinson. The poems all have common themes of personal transformation achieved through the introspective observations of the poet. An unusual chamber ensemble was chosen to include instruments not commonly used in vocal chamber music in order to create a greater variety of musical colors and timbres. This project included the creation of the musical score, a live performance that was video recorded, and the research paper. This document discusses the process of working with the composer, rehearsing the music as it was being composed, and negotiating revisions necessary to make the music more effective in performance. Each song is discussed in detail, especially the connection between the music and poetry, the overall form of the song, revisions discussed and implemented, and important motivic relationships between the songs that unify the cycle. In summary, the process of collaborating with a composer is a rewarding experience for both the performers and the composer, as everyone is challenged to improve their craft and overcome obstacles to achieve a successful performance.
ContributorsLoGiudice, Rosa Mia (Author) / Campbell, Andrew (Thesis advisor) / Ryan, Russell (Committee member) / Rockmaker, Jody (Committee member) / Arizona State University (Publisher)
Created2020