Detecting Malware Code as Video With Compressed, Time-Distributed Neural Networks

Santacroce, Michael; Koranek, Daniel; Jha, Rashmi

doi:10.1109/access.2020.3010706

Cited by 6 publications

(3 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In Reference 20, a new methodology for converting executable bytecode into a video, instead of image, is employed to classify alongside deep, time‐distributed neural networks that attain till 98.74% testing accuracy upon nine classes of malware, and till 99.36% testing accuracy upon balanced array of malicious versus benign files. Network can as well as classify the entire malware in the considered dataset for false‐positive rate of 13% and is noticed victorious too in classifying exclusively sections of an input besides early victory in a 0‐day situation.…”

Section: Related Studiesmentioning

confidence: 99%

An innovative malware detection methodology employing the amalgamation of stacked BiLSTM and CNN+LSTM‐based classification networks with the assistance of Mayfly metaheuristic optimization algorithm in cyber‐attack

Srinivasan

Deepalakshmi

2023

Concurrency and Computation

View full text Add to dashboard Cite

Summary The propagation of bias and its data have sensitive and critical security enterprises, which emphasizing the significance of perfecting state‐of‐the‐art intrusion discovery systems. These days malware pitfalls have turned precipitously vigorous, intricate and independently on this bias and its data. The artificial intelligence approaches have turned into the focus for cybersecurity as these are noticed acting as more applicable for dealing with contemporary malware events. Particularly, the neural networks having its robust conception prosecution which capability can challenge an expansive range of cyber pitfalls. This article portrays the advancement and testing of a neural network for bracket procedures. Also, the point election could be regarded as a combinatorial optimization issue. It also includes a new point selection fashion known as the Mayfly metaheuristic algorithm, which incorporates an S‐shaped transfer function for converting it into a double variant of the Mayfly Algorithm. While distinct seeker results were acquired out of different areas of the hunt space employing the Mayfly Algorithm, a lesser result could be assured. also, there are two bracket networks like piled BiLSTM and CNN LSTM that are regarded for the UNSW‐NB15 dataset, which automatically learn features out of the undressed data for landing the vicious train structure patterns and law sequence patterns. This largely lessens the price of artificial features engineering. The proffered bracket networks are examined employing the criteria like accuracy, precision, recall, F1‐score, and AUC score under binary and multi classes. It is observed that the proffered bracket networks achieve 91.87 of accuracy, 92.769 of precision, 90.47 of recall, and 91.6 of F1‐ score.

show abstract

Section: Related Studiesmentioning

confidence: 99%

An innovative malware detection methodology employing the amalgamation of stacked BiLSTM and CNN+LSTM‐based classification networks with the assistance of Mayfly metaheuristic optimization algorithm in cyber‐attack

Srinivasan

Deepalakshmi

2023

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…Tackling Non-Unique Redundancy Regularizing or pruning via similarity is a well-explored topic (Ayinde et al, 2019;Zhu et al, 2018;Srinivas & Babu, 2015;Santacroce et al, 2020). However, our approach integrate more cleanly with movement to insulate weights from regularization, with only a small increase in training time as listed in Appendix A.…”

Section: Globally Unique Movementmentioning

confidence: 99%

What Matters In The Structured Pruning of Generative Language Models?

Santacroce¹,

Wen²,

Shen³

et al. 2023

Preprint

View full text Add to dashboard Cite

Auto-regressive large language models such as GPT-3 require enormous computational resources to use. Traditionally, structured pruning methods are employed to reduce resource usage. However, their application to and efficacy for generative language models is heavily under-explored. In this paper we conduct an comprehensive evaluation of common structured pruning methods, including magnitude, random, and movement (Lagunas et al., 2021) pruning on the feed-forward layers in GPT-type models. Unexpectedly, randomly pruning results in performance that is comparable to the best established methods, across multiple natural language generation tasks. To understand these results, we provide a framework for measuring neuron-level redundancy of models pruned by different methods, and discover that established structured pruning methods do not take into account the distinctiveness of neurons, leaving behind excess redundancies. In view of this, we introduce Globally Unique Movement (GUM) to improve the uniqueness of neurons in pruned models. We then discuss the effects of our techniques on different redundancy metrics to explain the improved performance.

show abstract

“…Consequently, after division of the input into windows, each window is subjected to the same embeddings, special dropout, Bidirectional LSTM, and dense operations (Santacroce et al 2020). The rationale behind using a TimeDistributed output layer is to perform classi cation in each time step, enabling us to predict the error types of many words in an input sentence.…”

Section: Dense Layermentioning

confidence: 99%

Persian Typographical Error Type Detection Using Deep Neural Networks on Algorithmically-Generated Misspellings

Dehghani,

Faili

2023

Preprint

View full text Add to dashboard Cite

Spelling correction is a remarkable challenge in the field of natural language processing. The objective of spelling correction tasks is to recognize and rectify spelling errors automatically. The development of applications that can effectually diagnose and correct Persian spelling and grammatical errors has become more important in order to improve the quality of Persian text. The Typographical Error Type Detection in Persian is a relatively understudied area. Therefore, this paper presents a compelling approach for detecting typographical errors in Persian texts. Our work includes the presentation of a publicly available dataset called FarsTypo, which comprises 3.4 million words arranged in chronological order and tagged with their corresponding part-of-speech. These words cover a wide range of topics and linguistic styles. We develop an algorithm designed to apply Persian-specific errors to a scalable portion of these words, resulting in a parallel dataset of correct and incorrect words. By leveraging FarsTypo, we establish a strong foundation and conduct a thorough comparison of various methodologies employing different architectures. Additionally, we introduce a groundbreaking Deep Sequential Neural Network that utilizes both word and character embeddings, along with bidirectional LSTM layers, for token classification aimed at detecting typographical errors across 51 distinct classes. Our approach is contrasted with highly advanced industrial systems that, unlike this study, have been developed using a diverse range of resources. The outcomes of our final method proved to be highly competitive, achieving an accuracy of 97.62%, precision of 98.83%, recall of 98.61%, and surpassing others in terms of speed.

show abstract

Detecting Malware Code as Video With Compressed, Time-Distributed Neural Networks

Cited by 6 publications

References 16 publications

An innovative malware detection methodology employing the amalgamation of stacked BiLSTM and CNN+LSTM‐based classification networks with the assistance of Mayfly metaheuristic optimization algorithm in cyber‐attack

An innovative malware detection methodology employing the amalgamation of stacked BiLSTM and CNN+LSTM‐based classification networks with the assistance of Mayfly metaheuristic optimization algorithm in cyber‐attack

What Matters In The Structured Pruning of Generative Language Models?

Persian Typographical Error Type Detection Using Deep Neural Networks on Algorithmically-Generated Misspellings

Contact Info

Product

Resources

About