دانلود مقاله چالش های بلاک چین برای یادگیری عمیق
چکیده
1 مقدمه
2 پیش زمینه
3 طبقه بندی چارچوب های یادگیری عمیق مبتنی بر بلاک چین
4 چارچوب های یادگیری عمیق مبتنی بر بلاک چین
5 چالش ها و فرصت های تحقیق
6. نتیجه گیری
منابع
Abstract
1 Introduction
2 Background
3 Taxonomy of blockchain-based deep learning frameworks
4 Blockchain-based deep learning frameworks
5 Research challenges and opportunities
6 Conclusion
Declarations
References
چکیده
یادگیری عمیق به دلیل پتانسیل آن برای تصمیم گیری آگاهانه در سال های اخیر جذابیت زیادی پیدا کرده است. بخش بزرگی از سیستم‌های یادگیری عمیق امروزی مبتنی بر سرورهای متمرکز هستند و در ارائه شفافیت عملیاتی، قابلیت ردیابی، قابلیت اطمینان، امنیت و ویژگی‌های منشأ داده قابل اعتماد کوتاهی می‌کنند. همچنین، آموزش مدل‌های یادگیری عمیق با استفاده از داده‌های متمرکز در برابر مشکل تک نقطه‌ای آسیب‌پذیر است. در این مقاله، اهمیت ادغام فناوری بلاک چین با یادگیری عمیق را بررسی می کنیم. ما ادبیات موجود متمرکز بر ادغام بلاک چین با یادگیری عمیق را مرور می کنیم. ما ادبیات را با ابداع یک طبقه‌بندی موضوعی بر اساس هفت پارامتر طبقه‌بندی و دسته‌بندی می‌کنیم. یعنی نوع بلاک چین، مدل‌های یادگیری عمیق، پروتکل‌های اجماع خاص یادگیری عمیق، حوزه کاربردی، خدمات، انواع داده‌ها و اهداف استقرار. ما با برجسته کردن نقاط قوت و ضعف آن‌ها، بحث‌های روشنگرانه‌ای را درباره چارچوب‌های یادگیری عمیق مبتنی بر بلاک چین ارائه می‌کنیم. علاوه بر این، ما چارچوب‌های یادگیری عمیق مبتنی بر بلاک چین را بر اساس چهار پارامتر مانند نوع بلاک چین، پروتکل اجماع، روش یادگیری عمیق و مجموعه داده مقایسه می‌کنیم. در نهایت، چالش‌های تحقیقاتی مهمی را ارائه می‌کنیم که باید برای توسعه چارچوب‌های یادگیری عمیق بسیار قابل اعتماد مورد توجه قرار گیرند.
توجه! این متن ترجمه ماشینی بوده و توسط مترجمین ای ترجمه، ترجمه نشده است.
Abstract
Deep learning has gained huge traction in recent years because of its potential to make informed decisions. A large portion of today’s deep learning systems are based on centralized servers and fall short in providing operational transparency, traceability, reliability, security, and trusted data provenance features. Also, training deep learning models by utilizing centralized data is vulnerable to the single point of failure problem. In this paper, we explore the importance of integrating blockchain technology with deep learning. We review the existing literature focused on the integration of blockchain with deep learning. We classify and categorize the literature by devising a thematic taxonomy based on seven parameters; namely, blockchain type, deep learning models, deep learning specific consensus protocols, application area, services, data types, and deployment goals. We provide insightful discussions on the state-of-the-art blockchain-based deep learning frameworks by highlighting their strengths and weaknesses. Furthermore, we compare the existing blockchain-based deep learning frameworks based on four parameters such as blockchain type, consensus protocol, deep learning method, and dataset. Finally, we present important research challenges which need to be addressed to develop highly trustworthy deep learning frameworks.
Introduction
The potential of deep learning has been witnessed in almost all industrial sectors. For example, in the healthcare sector, deep learning models are used by physicians to correctly diagnose the disease of the patient from the symptoms. During the recent pandemic caused by the spread of coronavirus disease (COVID-19), deep learning models have been employed to predict the disease spread rate in a particular region and assist the authorities in managing the pandemic using the forecasted results [1,2,3]. Also, novel deep learning techniques have assisted health physicians in diagnosing COVID-19 patients using the dataset of CT and X-ray images [4, 5]. Apart from deep learning applications in the healthcare industry, it has been employed by security officers at airports to identify and verify banned items in passengers’ luggage or safeguarding software from vulnerabilities [6,7,8]. Using biometric security and face recognition features, deep learning models can assist the authorities in recognizing any physical dangers in real-time. The efficacy and efficiency of a deep learning system basis on the quality of the data used during the model training phase [9]. The majority of the deep learning techniques have considered centralized storage and processing for training the model that is prone to a single point of failure and data alteration by the adversaries. Any alteration of the data used for deep learning operations can corrupt the training model. Blockchain is a decentralized technology that can efficiently handle data integrity, security, and confidentiality [2, 10, 11, 11]. The integration of blockchain with deep learning can bring several benefits, e.g., automated and trusted decision making, efficient data market management, data security, better model building for prediction purposes, model sharing, and enhancement of the robustness of the deep learning-based systems.
Conclusion
In this paper, we have reviewed the state-of-the-art blockchain-based deep learning frameworks. We presented the key features of blockchain and deep learning along with a detailed discussion on the benefits resulted from their integration. The successful integration of deep learning with blockchain can facilitate in terms of data security and privacy to the existing systems and enhance the QoS in several applications mainly related to healthcare, blockchain security, data traffic management, and vehicular communication in urban areas. We devised a taxonomy to categorize the reported literature in several categories based on seven parameters such as blockchain type, deep learning models, deep learning specific consensus protocols, services, application areas, deployment goals, and data types. The critical aspects of existing blockchain-based deep learning frameworks are analyzed through a comprehensive analysis of the reported frameworks. Finally, we identified and discussed several technological and social challenges and barriers that require further research to unlock the full potential of blockchain in deep learning-based systems. Our concluding remarks along with the key recommendations include.