Time Complexity in Deep Learning Models

Shah, Bhoomi; Bhavsar, Hetal

doi:10.1016/j.procs.2022.12.023

Cited by 40 publications

(10 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In terms of dataset, number of samples, or size of the data in terms of GB or TB, number of output class labels. A demonstration on computing the time complexity associated with the convolutional model can be found in Reference 98. Results reveal that the designed model records the tradeoff between the model's accuracy versus running time.…”

Section: Resultsmentioning

confidence: 99%

An anomaly detection on blockchain infrastructure using artificial intelligence techniques: Challenges and future directions – A review

Chithanuru

Ramaiah

2023

Concurrency and Computation

View full text Add to dashboard Cite

SummaryRecently, Blockchain cryptographic distributed transaction ledger technology finds its usage in many applications. The application's ledgers implemented through Blockchain, ensures tamper‐proof transactions, and in turn the applications became robust enough against cyber‐attack But still adversaries put forward their efforts in detecting the vulnerabilities in the infrastructure to execute their ill intent. In the literature, many counter measures techniques are presented to address the security breaches on the Blockchain. Detecting as well mitigating from the possible anomalies against on blockchain infrastructure through AI techniques is the greatest attempt of this article, and which is much needed now. Hence, this review article enlightens the readers with the essence of cyber security, the security aspects of Blockchain, its infrastructure vulnerabilities, various Blockchain‐enabled use cases along with the their challenges. Primarily, anomaly detection on Blockchain infrastructure through Artificial Intelligence Techniques is focused. A detailed analysis of Artificial Intelligence Techniques in detecting the anomalies with the help of Blockchain and also how these two technologies complement each other was demonstrated with the help of suitable use cases. The merits, challenges along with the possible future directions, while integrating Blockchain with Artificial Intelligence Techniques are presented for the benefit of research community.

show abstract

Section: Resultsmentioning

confidence: 99%

An anomaly detection on blockchain infrastructure using artificial intelligence techniques: Challenges and future directions – A review

Chithanuru

Ramaiah

2023

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…In this matter, it has been demonstrated through early trials that are incorporating a retrainable encoder as the primary layer of an advanced HAR classifier results in achieving the highest level of performance. In order to justify its computational cost, it is essential to demonstrate that the superior performance of the network is not solely due to its larger structure or the presence of more trainable parameters [ 28 ]. Please note that when an encoder layer is added to the classifier, and it requires retraining, the number of trainable parameters will increase, despite the increase in performance.…”

Section: System Methodsmentioning

confidence: 99%

CSI-Based Human Activity Recognition Using Multi-Input Multi-Output Autoencoder and Fine-Tuning

Chahoushi

Nabati

Asvadi

et al. 2023

Sensors

View full text Add to dashboard Cite

Wi-Fi-based human activity recognition (HAR) has gained considerable attention recently due to its ease of use and the availability of its infrastructures and sensors. Channel state information (CSI) captures how Wi-Fi signals are transmitted through the environment. Using channel state information of the received signals transmitted from Wi-Fi access points, human activity can be recognized with more accuracy compared with the received signal strength indicator (RSSI). However, in many scenarios and applications, there is a serious limit in the volume of training data because of cost, time, or resource constraints. In this study, multiple deep learning models have been trained for HAR to achieve an acceptable accuracy level while using less training data compared to other machine learning techniques. To do so, a pretrained encoder which is trained using only a limited number of data samples, is utilized for feature extraction. Then, by using fine-tuning, this encoder is utilized in the classifier, which is trained by a fraction of the rest of the data, and the training is continued alongside the rest of the classifier’s layers. Simulation results show that by using only 50% of the training data, there is a 20% improvement compared with the case where the encoder is not used. We also showed that by using an untrainable encoder, an accuracy improvement of 11% using 50% of the training data is achievable with a lower complexity level.

show abstract

“…Two other factors that are sensitive and have an impact on computational complexity are batch size

(b)

and learning rate

\left(\eta \right)

, according to study findings these two parameters must be multiplied 58 . So, after considering batch size and learning rate, overall time complexity can be defined as:

$$ \left(\mathcal{O}\left(n.k.

…”

Section: Proposed Modelmentioning

confidence: 99%

Efficient breast cancer diagnosis using multi‐level progressive feature aggregation based deep transfer learning system

Patel,

Chaurasia

2024

Int J Imaging Syst Tech

View full text Add to dashboard Cite

Breast cancer is a worldwide fatal disease that exists mostly among women. The deep learning technique has proven its effectiveness, but the performance of the existing deep learning systems is quite compromising. In this work, a deep transfer learning system is suggested for efficient breast cancer classification from histopathology images. This system is based on a novel multi‐level progressive feature aggregation (MPFA) and a spatial domain learning approach. The combination of a pretrained Resnet101 backbone network with MPFA is implemented to extract more significant features. In addition, a mixed‐dilated spatial domain learning network (MSLN) is further incorporated to enhance the receptive field and increase discrimination between features. The proposed method achieved superior performance as compared to the existing state‐of‐the‐art methods, offering 99.24% accuracy, a 98.79% F‐1 score, 98.59% precision, and 98.99% recall values over BreaKHis dataset. An ablation study is carried out over the ICIAR2018 dataset to verify the generalizability and effectiveness of the system.

show abstract

Time Complexity in Deep Learning Models

Cited by 40 publications

References 10 publications

An anomaly detection on blockchain infrastructure using artificial intelligence techniques: Challenges and future directions – A review

An anomaly detection on blockchain infrastructure using artificial intelligence techniques: Challenges and future directions – A review

CSI-Based Human Activity Recognition Using Multi-Input Multi-Output Autoencoder and Fine-Tuning

Efficient breast cancer diagnosis using multi‐level progressive feature aggregation based deep transfer learning system

Contact Info

Product

Resources

About