182 research outputs found
sj-pdf-1-vmj-10.1177_1358863X221094082 – Supplemental material for Artificial intelligence for the evaluation of peripheral artery disease using arterial Doppler waveforms to predict abnormal ankle-brachial index
No full textSupplemental material, sj-pdf-1-vmj-10.1177_1358863X221094082 for Artificial intelligence for the evaluation of peripheral artery disease using arterial Doppler waveforms to predict abnormal ankle-brachial index by Robert D McBane, Dennis H Murphree, David Liedl, Francisco Lopez-Jimenez, Itzhak Zachi Attia, Adelaide Arruda-Olson, Christopher G Scott, Naresh Prodduturi, Steve E Nowakowski, Thom W Rooke, Ana I Casanegra, Waldemar E Wysokinski, Keith E Swanson, Damon E Houghton, Haraldur Bjarnason and Paul W Wennberg in Vascular Medicine</p
Essentials of Marketing Research : A Hands-On Orientation
No full textFor courses in Marketing Research at two- and four-year colleges and universities An engaging, do-it-yourself approach to marketing research Essentials of Marketing Research: A Hands-On Orientation presents a concise overview of marketing research via a do-it-yourself approach that engages students. Building on the foundation of his successful previous titles-Basic Marketing Research: Integration of Social Media and Marketing Research: An Applied Orientation-author Naresh Malhotra covers concepts at an elementary level, deemphasizing statistics and formulas. Sensitive to the needs of today\u27s undergraduates, Malhotra integrates online and social media content, and provides current, contemporary examples that ground course material in the real world
Author Correction: UClncR: Ultrafast and comprehensive long non-coding RNA detection from RNA-seq
No full textA correction to this article has been published and is linked from the HTML and PDF versions of this paper. The error has not been fixed in the paper.</jats:p
Low-complexity convolutional neural networks for automatic target recognition
No full textOver the decades, several algorithms have been proposed for designing automatic target recognition systems based on synthetic aperture radar imagery. Recently, with the rise of Deep Learning, there has been growing interest in developing neural network based automatic target recognition systems for synthetic aperture radar applications. However, these networks are typically complex in terms of storage and computation which inhibits their deployment in the field, where such resources are heavily constrained.
In order to reduce the cost of implementing these networks, in this thesis we develop a set of compact network architectures and train them in fixed-point. Our proposed method achieves an overall 984× reduction in terms of storage requirements and 71× reduction in terms of computational complexity compared to state-of-the-art convolutional neural networks for automatic target recognition, while maintaining a classification accuracy of >99% on the MSTAR dataset.Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2022-05-01The student, Hassan Dbouk, accepted the attached license on 2020-03-10 at 11:07.The student, Hassan Dbouk, submitted this Thesis for approval on 2020-03-10 at 11:18.This Thesis was approved for publication on 2020-03-11 at 11:49.DSpace SAF Submission Ingestion Package generated from Vireo submission #14893 on 2020-08-25 at 17:26:59Made available in DSpace on 2020-08-26T23:51:24Z (GMT). No. of bitstreams: 2
DBOUK-THESIS-2020.pdf: 1264282 bytes, checksum: 081816a0d9fac969b3084e01e987d955 (MD5)
LICENSE.txt: 4209 bytes, checksum: 555c4643e121e7aa3736b7cfffe3f5f4 (MD5)
Previous issue date: 2020-03-11Embargo set by: Seth Robbins for item 115700
Lift date: 2022-08-26T23:51:32Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemEmbargo set by: Seth Robbins for item 115700
Lift date: 2022-08-26T23:54:40Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemEmbargo set by: Seth Robbins for item 115700
Lift date: 2022-08-26T23:55:59Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemEmbargo set by: Seth Robbins for item 115700
Lift date: 2022-08-26T23:57:28Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemEmbargo set by: Seth Robbins for item 115700
Lift date: 2022-08-26T23:58:55Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemAuthor requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemU of I Onl
Structural and optical properties of tin disulphide thin films grown by flash evaporation
No full textAesoper Galpa Samagra (Aesop's Fables: Bengali)
No full textHere is a genuine find in this sense: I found it without help on the shelves in this fine bookshop near St. Xavier's College in Kolkata. The picture on the cover, including a fox and some grapes, gave me the clue. It is my first book exclusively in Bengali; I have a bilingual copy of Fox Fables that has both Bengali and English. This book features about a dozen full-page black-and-white designs. Among the best of them are The Ethiopian and CW. I cannot give page numbers because those too are in Bengali! I am proud of finding this book on my busy day in Kolkata!This is a hardbound book (hard cover)Language note: BengaliRetold by Naresh Chandra Jan
In-sensor information processing for resource-limited platforms on flexible epidermal substrates
No full textMoving towards the age of big data, the demand for embedded processing has been drastically increasing to make inference and intelligent decisions at lower architectural layers. The myriad of health conditions that can be treated and analyzed via low-energy embedded information processing kernels drives the demand for biomedical circuits, with optimized performance and cost. A large class of these healthcare applications require digital signal processing algorithms to be implemented with strict resources, such as energy and silicon area. Shrinking technology nodes produce both higher computing performance and energy efficiency. However, energy delivery and communication circuitry have not benefited significantly from technology scaling due to different sets of figures of merit. In-sensor information processing can be utilized to lower the energy consumption of such systems by eliminating the redundant volume of data traffic between the sensors and the central processing station. This work focuses on embedding intelligence on the epidermal flexible substrates to extract and analyze critical biomedical information for in-situ diagnosis. The primary objective of this work is illustrating the advantages of epidermal electronics combined with robust information processing systems, at system and application level. The major challenge is the design of robust and efficient algorithms for reliable operation on resource limited hardware platforms and flexible substrate non-idealities. To do so, we developed the first in-sensor ECG and PPG processors on flexible epidermal substrates. The systems are first prototyped using discrete components, followed by an ASIC implementation. Measurement results show that the in-sensor information processing has reduced the transmitted data traffic by 150X, and the system energy consumption by 3.56X.Submission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2018-12-01The student, Pourya Assem, accepted the attached license on 2016-07-05 at 17:47.The student, Pourya Assem, submitted this Thesis for approval on 2016-07-05 at 17:49.This Thesis was approved for publication on 2016-07-07 at 12:24.DSpace SAF Submission Ingestion Package generated from Vireo submission #9769 on 2017-02-28 at 14:40:24Made available in DSpace on 2017-03-01T17:02:15Z (GMT). No. of bitstreams: 2
ASSEM-THESIS-2016.pdf: 46560117 bytes, checksum: f1fd6d103ad2cf5d41428429e3c4a997 (MD5)
LICENSE.txt: 4209 bytes, checksum: b8f8e41401df07a6b2468fc7a3dc7553 (MD5)
Previous issue date: 2016-07-07Embargo set by: Seth Robbins for item 98744
Lift date: 2019-03-01T17:02:22Z
Reason: Author requested closed access (OA after 2yrs) in Vireo ETD systemEmbargo set by: Seth Robbins for item 98744
Lift date: 2019-03-01T17:03:32Z
Reason: Author requested closed access (OA after 2yrs) in Vireo ETD systemEmbargo set by: Seth Robbins for item 98744
Lift date: 2019-03-01T17:05:02Z
Reason: Author requested closed access (OA after 2yrs) in Vireo ETD systemEmbargo set by: Seth Robbins for item 98744
Lift date: 2019-03-01T17:06:55Z
Reason: Author requested closed access (OA after 2yrs) in Vireo ETD systemLimited Restriction Lifted for Item 98744 on 2019-03-02T10:15:27Z
Discrete state-space active inference in nonstationary environments
No full textActive inference is a neuroscientific theory, which states that all living systems (e.g. the human brain) minimize a quantity termed the free energy. By minimizing this free energy, living systems keep an accurate representation of the world in their internal model (learning), are provided with an optimal way of acting on the world (action selection), and can predict incoming sensory data (perception). Considering the fact that these properties are sought after in artificial intelligence systems as well, active inference has also become an interesting topic from an engineering point of view. The application of active inference can be done with both a continuous and a discrete state-space framework. However, research on discrete state-space active inference has neglected the extension of its applicability to nonstationary environments. This work aims to fill that gap. More specifically, the goal of this research is to evaluate the performance of state-of-the-art discrete state-space active inference agents in nonstationary environments, and assess whether forgetting part of the agent's previous experiences can increase its performance. The type of nonstationarity that is used in this work is cyclostationarity, and this nonstationarity will only be manifested in the transition process of the active inference task. Moreover, the specific type of task solved is one of planning and navigation in a gridworld. Since the agent has to deal with a planning and navigation task, performance is quantified by the number of steps the agent needs to take in order to reach its goal. Three methods of forgetting are implemented and compared, inspired by techniques from reinforcement learning, deep learning and time series analysis respectively. These are: (1) the use of a constant forget rate, (2) the use of the updating mechanism of a long short-term memory (LSTM) cell applied to the updating of the generative model in active inference, and (3) the use of a memory window that stores experience only from a certain trial onwards and forgets experience from before this trial by utilizing a rolling summation of the concentration parameters. The results show that forgetting with the use of a memory window can significantly improve performance, provided that the agent can reach the goal state from the initial state in one trial. When this is not the case, the use of a memory window does not (positively or negatively) influence performance. Both the implementation of forgetting based on the updating of an LSTM cell and the use of a constant forget rate have unanimously shown to decrease performance, and thus should not be implemented in active inference.Mechanical Engineerin
- …
