2018/05/21 by John Clemens · 1 voice
Computer Science · Mathematics · #Adversarial Robustness in Machine Learning #Machine Learning and Algorithms #Software Testing and Debugging Techniques #cs.LG #stat.ML
paper · pdf · doi:10.1109/ijcnn.2018.8489466
Under review for publication at IJCNN 2018
arxiv created 2018/05/21 · arxiv published 2018/05/21 · openalex created_date 2018/06/01 · openalex publication_date 2018/07/01 · arxiv updated 2018/10/23 · openalex updated_date 2026/07/29
Recurrent neural networks (RNNs) are powerful constructs capable of modeling complex systems, up to and including Turing Machines. However, learning such complex models from finite training sets can be difficult. In this paper we empirically show that RNNs can learn models of computer peripheral devices through input and output state observation. This enables automated development of functional software-only models of hardware devices. Such models are applicable to any number of tasks, including device validation, driver development, code de-obfuscation, and reverse engineering. We show that the same RNN structure successfully models six different devices from simple test circuits up to a 16550 UART serial port, and verify that these models are capable of producing equivalent output to real hardware.