Matching Items (153)
Filtering by
- All Subjects: Electrical Engineering
- Creators: Karady, George G.
- Creators: Cao, Yu
- Resource Type: Text
Description
Artificial intelligence is one of the leading technologies that mimics the problem solving and decision making capabilities of the human brain. Machine learning algorithms, especially deep learning algorithms, are leading the way in terms of performance and robustness. They are used for various purposes, mainly for computer vision, speech recognition, and object detection. The algorithms are usually tested inaccuracy, and they utilize full floating-point precision (32 bits). The hardware would require a high amount of power and area to accommodate many parameters with full precision. In this exploratory work, the convolution autoencoder is quantized for the working of an event base camera. The model is designed so that the autoencoder can work on-chip, which would sufficiently decrease the latency in processing. Different quantization methods are used to quantize and binarize the weights and activations of this neural network model to be portable and power efficient. The sparsity term is added to make the model as robust and energy-efficient as possible. The network model was able to recoup the lost accuracy due to binarizing the weights and activation's to quantize the layers of the encoder selectively. This method of recouping the accuracy gives enough flexibility to introduce the network on the chip to get real-time processing from systems like event-based cameras. Lately, computer vision, especially object detection have made strides in their object detection accuracy. The algorithms can sufficiently detect and predict the objects in real-time. However, end-to-end detection of the algorithm is challenging due to the large parameter need and processing requirements. A change in the Non Maximum Suppression algorithm in SSD(Single Shot Detector)-Mobilenet-V1 resulted in less computational complexity without change in the quality of output metric. The Mean Average Precision(mAP) calculated suggests that this method can be implemented in the post-processing of other networks.
ContributorsKuzhively, Ajay Balu (Author) / Cao, Yu (Thesis advisor) / Seo, Jae-Sun (Committee member) / Fan, Delian (Committee member) / Arizona State University (Publisher)
Created2021
Description
Articial Neural Network(ANN) has become a for-bearer in the field of Articial Intel-
ligence. The innovations in ANN has led to ground breaking technological advances
like self-driving vehicles,medical diagnosis,speech Processing,personal assistants and
many more. These were inspired by evolution and working of our brains. Similar
to how our brain evolved using a combination of epigenetics and live stimulus,ANN
require training to learn patterns.The training usually requires a lot of computation
and memory accesses. To realize these systems in real embedded hardware many
Energy/Power/Performance issues needs to be solved. The purpose of this research
is to focus on methods to study data movement requirement for generic Neural Net-
work along with the energy associated with it and suggest some ways to improve the
design.Many methods have suggested ways to optimize using mix of computation and
data movement solutions without affecting task accuracy. But these methods lack a
computation model to calculate the energy and depend on mere back of the envelope calculation. We realized that there is a need for a generic quantitative analysis
for memory access energy which helps in better architectural exploration. We show
that the present architectural tools are either incompatible or too slow and we need
a better analytical method to estimate data movement energy. We also propose a
simplistic yet effective approach that is robust and expandable by users to support
various systems.
ligence. The innovations in ANN has led to ground breaking technological advances
like self-driving vehicles,medical diagnosis,speech Processing,personal assistants and
many more. These were inspired by evolution and working of our brains. Similar
to how our brain evolved using a combination of epigenetics and live stimulus,ANN
require training to learn patterns.The training usually requires a lot of computation
and memory accesses. To realize these systems in real embedded hardware many
Energy/Power/Performance issues needs to be solved. The purpose of this research
is to focus on methods to study data movement requirement for generic Neural Net-
work along with the energy associated with it and suggest some ways to improve the
design.Many methods have suggested ways to optimize using mix of computation and
data movement solutions without affecting task accuracy. But these methods lack a
computation model to calculate the energy and depend on mere back of the envelope calculation. We realized that there is a need for a generic quantitative analysis
for memory access energy which helps in better architectural exploration. We show
that the present architectural tools are either incompatible or too slow and we need
a better analytical method to estimate data movement energy. We also propose a
simplistic yet effective approach that is robust and expandable by users to support
various systems.
ContributorsChowdary, Hidayatullah (Author) / Cao, Yu (Thesis advisor) / Seo, JaeSun (Committee member) / Chakrabarti, Chaitali (Committee member) / Arizona State University (Publisher)
Created2018
Description
The inherent intermittency in solar energy resources poses challenges to scheduling generation, transmission, and distribution systems. Energy storage devices are often used to mitigate variability in renewable asset generation and provide a mechanism to shift renewable power between periods of the day. In the absence of storage, however, time series forecasting techniques can be used to estimate future solar resource availability to improve the accuracy of solar generator scheduling. The knowledge of future solar availability helps scheduling solar generation at high-penetration levels, and assists with the selection and scheduling of spinning reserves. This study employs statistical techniques to improve the accuracy of solar resource forecasts that are in turn used to estimate solar photovoltaic (PV) power generation. The first part of the study involves time series forecasting of the global horizontal irradiation (GHI) in Phoenix, Arizona using Seasonal Autoregressive Integrated Moving Average (SARIMA) models. A comparative study is completed for time series forecasting models developed with different time step resolutions, forecasting start time, forecasting time horizons, training data, and transformations for data measured at Phoenix, Arizona. Approximately 3,000 models were generated and evaluated across the entire study. One major finding is that forecasted values one day ahead are near repeats of the preceding day—due to the 24-hour seasonal differencing—indicating that use of statistical forecasting over multiple days creates a repeating pattern. Logarithmic transform data were found to perform poorly in nearly all cases relative to untransformed or square-root transform data when forecasting out to four days. Forecasts using a logarithmic transform followed a similar profile as the immediate day prior whereas forecasts using untransformed and square-root transform data had smoother daily solar profiles that better represented the average intraday profile. Error values were generally lower during mornings and evenings and higher during midday. Regarding one-day forecasting and shorter forecasting horizons, the logarithmic transformation performed better than untransformed data and square-root transformed data irrespective of forecast horizon for data resolutions of 1-hour, 30-minutes, and 15-minutes.
ContributorsSoundiah Regunathan Rajasekaran, Dhiwaakar Purusothaman (Author) / Johnson, Nathan G (Thesis advisor) / Karady, George G. (Thesis advisor) / Ayyanar, Raja (Committee member) / Arizona State University (Publisher)
Created2016