Applying Genetic Programming to Bytecode and Assembly

Traditional genetic programming (GP) is typically not used to perform unrestricted evolution on entire programs at the source code level. Instead, only small sections within pro-grams are usually evolved. Not being able to evolve whole programs is an issue since it limits the flexibility of what can be evolved. Evolving programs in either bytecode or as-sembly language is a method that has been used to perform unrestricted evolution. This paper provides an overview of applying genetic programming to Java bytecode and x86 as-sembly. Two examples of how this method has been imple-mented will be explored. We will also discuss experimental results that include evolving recursive functions and auto-mated bug repair. Keywords evolutionary computation, x86 assembly code, Java byte-code, FINCH, automated bug repair 1