Avi Pfeffer
YOU?
Author Swipe
Simulation Intelligence: Towards a New Generation of Scientific Methods Open
The original "Seven Motifs" set forth a roadmap of essential methods for the field of scientific computing, where a motif is an algorithmic method that captures a pattern of computation and data movement. We present the "Nine Motifs of Sim…
Unifying AI Algorithms with Probabilistic Programming using Implicitly Defined Representations Open
We introduce Scruff, a new framework for developing AI systems using probabilistic programming. Scruff enables a variety of representations to be included, such as code with stochastic choices, neural networks, differential equations, and …
Software Adaptation for an Unmanned Undersea Vehicle Open
Most current software systems are not adaptable, making them less capable of achieving their objectives. We are developing a method to optimize software for new environments automatically. An independent evaluation has demonstrated that th…
Learning Probabilistic Programs Using Backpropagation Open
Probabilistic modeling enables combining domain knowledge with learning from data, thereby supporting learning from fewer training instances than purely data-driven methods. However, learning probabilistic models is difficult and has not a…
View article: Artificial Intelligence Based Malware Analysis
Artificial Intelligence Based Malware Analysis Open
Artificial intelligence methods have often been applied to perform specific functions or tasks in the cyber-defense realm. However, as adversary methods become more complex and difficult to divine, piecemeal efforts to understand cyber-att…
Structured Factored Inference: A Framework for Automated Reasoning in Probabilistic Programming Languages Open
Reasoning on large and complex real-world models is a computationally difficult task, yet one that is required for effective use of many AI applications. A plethora of inference algorithms have been developed that work well on specific mod…
Probabilistic Programming for Malware Analysis Open
Constructing lineages of malware is an important cyber-defense task. Performing this task is difficult, however, due to the amount of malware data and obfuscation techniques by the authors. In this work, we formulate the lineage task as a …