Deep Learning and Neural Networks

Mukhopadhyay, Sayan

doi:10.1007/978-1-4842-3450-1_5

Cited by 8 publications

(2 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since every model's input has a different range of value scale, the collected data in each input attribute have been pre-processed by using the z-score technique since it is a popular technique [38]. The aim of using this technique is to simplify the procedure of model training.…”

Section: ) Input Selectionmentioning

confidence: 99%

Agnostic Energy Consumption Models for Heterogeneous GPUs in Cloud Computing

Alnori,

Djemame,

Alsenani

2024

Applied Sciences

View full text Add to dashboard Cite

The adoption of cloud computing has grown significantly among individuals and in organizations. According to this growth, Cloud Service Providers have continuously expanded and updated cloud-computing infrastructures, which have become more heterogeneous. Managing these heterogeneous resources in cloud infrastructures while ensuring Quality of Service (QoS) and minimizing energy consumption is a prominent challenge. Therefore, unifying energy consumption models to deal with heterogeneous cloud environments is essential in order to efficiently manage these resources. This paper deeply analyzes factors affecting power consumption and employs these factors to develop power models. Because of the strong correlation between power consumption and energy consumption, the influencing factors on power consumption, with the addition of other factors, are considered when developing energy consumption models to enhance the treatment in heterogeneous infrastructures in cloud computing. These models have been developed for two Virtual Machines (VMs) containing heterogeneous Graphics Processing Units (GPUs) architectures with different features and capabilities. Experiments evaluate the models through a cloud testbed between the actual and predicted values produced by the models. Deep Neural Network (DNN) power models are validated with shallow neural networks using performance counters as inputs. Then, the results are significantly enhanced by 8% when using hybrid inputs (performance counters, GPU and memory utilization). Moreover, a DNN energy-agnostic model to abstract the complexity of heterogeneous GPU architectures is presented for the two VMs. A comparison between the standard and agnostic energy models containing common inputs is conducted in each VM. Agnostic energy models with common inputs for both VMs show a slight enhancement in accuracy with input reduction.

show abstract

Section: ) Input Selectionmentioning

confidence: 99%

Agnostic Energy Consumption Models for Heterogeneous GPUs in Cloud Computing

Alnori,

Djemame,

Alsenani

2024

Applied Sciences

View full text Add to dashboard Cite

show abstract

“…However, the algorithm requires a high accuracy of frequency and amplitude estimation. In addition, inspired by the working mechanism of the brain, deep neural networks [10]- [14] are designed to extract high-level feature information, which has achieved great success in many fields and specific tasks [15]- [18] such as, speech recognition [19], [20], image processing [21], natural language processing [22], and computer vision [23]. In the field of communication, signals are considered to be temporal data, which can be learned with deep learning to recognize its patterns inside [24].…”

Section: Introductionmentioning

confidence: 99%

Mixing Position Estimation of AIS Mixed Signal Based on Double-Sliding Window Method

Zhang

Meng

et al. 2019

IEEE Access

View full text Add to dashboard Cite

With the development of the shipping industry, Automatic Identification System (AIS) used for ship communication becomes more and more important. Aiming at the problem of mixing position estimation of AIS mixed signals, this paper improves the double sliding-window detection algorithm to estimate the mixing position of miscellaneous signals accurately. There is a significant difference in energy between the aliased and unmixed parts of the mixed signal. When unmixed parts just enter one of the energy detections windows, the decision function reaches its peak value by establishing a proper decision function, that is to say, the position of the beginning and end of the mixing part is estimated. The simulation results show that the proposed algorithm which is compared with the frequency and amplitude detection algorithm can achieve the mixing position estimation with low complexity and strong robustness, and the estimation accuracy is close to the Cramer-Rao Bound under the condition of the high signal-to-noise ratio.

show abstract

Deep Learning-Based Recommendation Systems: Review and Critical Analysis

Alam,

Ahmed

2024

Lecture Notes in Networks and Systems

View full text Add to dashboard Cite

Deep Learning and Neural Networks

Cited by 8 publications

References 3 publications

Agnostic Energy Consumption Models for Heterogeneous GPUs in Cloud Computing

Agnostic Energy Consumption Models for Heterogeneous GPUs in Cloud Computing

Mixing Position Estimation of AIS Mixed Signal Based on Double-Sliding Window Method

Deep Learning-Based Recommendation Systems: Review and Critical Analysis

Contact Info

Product

Resources

About