Results of the summarizing process for the arXiv paper: 1807.05620v4

NEUZZ is a new fuzzing technique developed to efficiently find software vulnerabilities. Fuzzing is a popular technique used to generate inputs that can trigger different bugs in software. However, existing state-of-the-art fuzzers are not very efficient at finding hard-to-trigger software bugs. Most popular fuzzers use evolutionary guidance to generate inputs but such algorithms often get stuck in fruitless sequences of random mutations. Gradient-guided optimization presents an alternative to evolutionary guidance and has been shown to outperform evolutionary algorithms in domains like machine learning by utilizing gradients or higher-order derivatives of the underlying function. However, gradient-guided approaches are not directly applicable to fuzzing as real-world program behaviors contain many discontinuities, plateaus and ridges where the gradient-based methods often get stuck. To address this problem, NEUZZ proposes a novel program smoothing technique using surrogate neural network models that can incrementally learn smooth approximations of a complex, real-world program's branching behaviors. This approach creates a smooth surrogate function approximating the discrete branching behavior of target programs and enables gradient-guided input generation schemes to improve the fuzzing efficiency. The extensive evaluations demonstrate that NEUZZ significantly outperforms 10 state-of-the art graybox fuzzers on 10 real world programs both at finding new bugs and achieving higher edge coverage. NEUZZ found 31 unknown bugs that other fuzzers failed to find in 10 real world programs and achieved 3X more edge coverage than all of the tested graybox fuzzers for 24 hours running. The authors behind NEUZZ include Dongdong She, Kexin Pei, Dave Epstein, Junfeng Yang, Baishakhi Ray and Suman Jana who presented their findings at the 40th IEEE Symposium on Security and Privacy held from May 20 - 22 in San Francisco.

• - NEUZZ is a new fuzzing technique developed to efficiently find software vulnerabilities.
• - Fuzzing is used to generate inputs that can trigger different bugs in software.
• - Existing state-of-the-art fuzzers are not very efficient at finding hard-to-trigger software bugs.
• - Gradient-guided optimization has been shown to outperform evolutionary algorithms in domains like machine learning by utilizing gradients or higher-order derivatives of the underlying function.
• - However, gradient-guided approaches are not directly applicable to fuzzing as real-world program behaviors contain many discontinuities, plateaus and ridges where the gradient-based methods often get stuck.
• - NEUZZ proposes a novel program smoothing technique using surrogate neural network models that can incrementally learn smooth approximations of a complex, real-world program's branching behaviors.
• - This approach creates a smooth surrogate function approximating the discrete branching behavior of target programs and enables gradient-guided input generation schemes to improve the fuzzing efficiency.
• - NEUZZ significantly outperforms 10 state-of-the art graybox fuzzers on 10 real world programs both at finding new bugs and achieving higher edge coverage.
• - NEUZZ found 31 unknown bugs that other fuzzers failed to find in 10 real world programs and achieved 3X more edge coverage than all of the tested graybox fuzzers for 24 hours running.
• - The authors behind NEUZZ include Dongdong She, Kexin Pei, Dave Epstein, Junfeng Yang, Baishakhi Ray and Suman Jana who presented their findings at the 40th IEEE Symposium on Security and Privacy held from May 20 - 22 in San Francisco.

NEUZZ is a new way to find problems in computer programs. Fuzzing is when you try different things to see if they cause problems in the program. NEUZZ is better at finding tricky problems than other fuzzing methods. Gradient-guided optimization is a way to make things work better by using math. But it doesn't always work for fuzzing because programs can be too complicated. NEUZZ uses a special technique with neural networks to make fuzzing more efficient and find more problems. The people who made NEUZZ presented their work at a conference called IEEE Symposium on Security and Privacy." Definitions- Fuzzing: trying different inputs to see if they cause problems in software - Surrogate neural network models: mathematical models that can learn how a program works and predict what will happen next - Gradient-guided optimization: using math to improve performance by following the direction of the slope of a function - Discontinuities, plateaus, ridges: parts of a program where the behavior changes suddenly or stays the same for a while before changing again - Graybox fuzzers: tools that test software by partially knowing how it works

Introducing NEUZZ: A Novel Fuzzing Technique for Efficiently Finding Software Vulnerabilities

Fuzzing is a popular technique used to generate inputs that can trigger different bugs in software. However, existing state-of-the-art fuzzers are not very efficient at finding hard-to-trigger software bugs. To address this problem, researchers from Columbia University have developed a novel fuzzing technique called NEUZZ (Neural Network Enhanced Unconstrained ZZ) which utilizes program smoothing and gradient-guided optimization to significantly improve the efficiency of fuzzing. The authors presented their findings at the 40th IEEE Symposium on Security and Privacy held from May 20 - 22 in San Francisco.

What is Fuzzing?

Fuzzing is a type of automated testing that involves providing invalid or unexpected data as input to a computer system or application with the goal of uncovering security vulnerabilities or other types of bugs. It works by generating random data and feeding it into an application or system in order to test its behavior under various conditions. The idea behind fuzzing is that if enough random data is provided, eventually some combination will cause the application or system to crash due to an unforeseen bug. This can then be used by developers to identify and fix potential security issues before they become exploitable by malicious actors.

Limitations of Existing Fuzzers

Most popular fuzzers use evolutionary guidance algorithms which often get stuck in fruitless sequences of random mutations when trying to find hard-to-trigger software bugs. Gradient-guided optimization presents an alternative but such approaches are not directly applicable to fuzzing as real world program behaviors contain many discontinuities, plateaus and ridges where gradient based methods often get stuck.

NEUZZ: Neural Network Enhanced Unconstrained ZZ

To address these limitations, NEUZZ proposes a novel program smoothing technique using surrogate neural network models that can incrementally learn smooth approximations of complex real world programs’ branching behaviors. This approach creates a smooth surrogate function approximating the discrete branching behavior of target programs and enables gradient guided input generation schemes which significantly improves the efficiency of fuzzing compared with existing techniques like evolutionary guidance algorithms.

Evaluation Results

The extensive evaluations demonstrate that NEUZZ significantly outperforms 10 state-of-the art graybox fuzzers on 10 real world programs both at finding new bugs and achieving higher edge coverage than all other tested graybox fuzzers within 24 hours running time period . Specifically, NEUZZ found 31 unknown bugs that other tested fuzzy failed to find in 10 real world programs while achieving 3X more edge coverage than all tested graybox fuzzy within 24 hours running time period .

Conclusion

In conclusion, this research paper introduces NEUZZ – a novel neural network enhanced unconstrained zz algorithm for efficiently finding software vulnerabilities through improved program smoothing and gradient guided optimization techniques compared with traditional evolutionary guidance algorithms used by most popular existing fuzzy tools today . The evaluation results show significant improvements over state -of -the art grey box fuzzy tools both at discovering new unknown bugs as well as achieving higher edge coverage within 24 hours running time period .