Multi-Class classification of vulnerabilities in smart contracts using AWD-LSTM, with pre-trained encoder inspired from natural language processing

Abstract

Abstract Vulnerability detection and safety of smart contracts are of paramount importance because of their immutable nature. Symbolic tools like OYENTE and MAIAN are typically used for vulnerability prediction in smart contracts. As these tools are computationally expensive, they are typically used to detect vulnerabilities until some predefined invocation depth. These tools require more search time as the invocation depth increases. Since the use of smart contracts increases rapidly, their analysis becomes difficult using these traditional tools. Recently, a machine learning technique called Long Short Term Memory (LSTM) has been used to predict the vulnerability of a smart contract. In the present article, we present how to classify smart contracts into Suicidal, Prodigal, Greedy, or Normal categories using Average Stochastic Gradient Descent Weight-Dropped LSTM (AWD-LSTM), a variant of LSTM. We reduced the class imbalance by considering only distinct opcode combinations for normal contracts and achieved a weighted average F1 score of 90.0%. Such techniques can be utilized in real-time to analyze a large number of smart contracts and to improve their security.