Location
https://www.kennesaw.edu/ccse/events/computing-showcase/fa25-cday-program.php
Document Type
Event
Start Date
24-11-2025 4:00 PM
Description
A neuron with binary inputs & outputs corresponds to a Boolean function. To explain and verify the behavior a neuron (and by extension, a neural network), we can explain and verify its Boolean function. There has been recent interest in representing the Boolean function of such a neuron as an Ordered Binary Decision Diagram (OBDD), which facilitates such analyses. We propose an algorithm for compiling a binary neuron into an OBDD using a compiler that decomposes a Boolean function into a decision graph. We augment this compiler so that it outputs an OBDD instead. Our augmented compiler produces intermediate OBDDs that represent inner- and outer-bounds of the original neuron, tightening compilation progresses. Theoretically, decision graphs of binary neurons are more succinct than their decision trees. Empirically, compilation to decision graphs can scale to neurons with over a thousand features, compared to dozens of features using other compilers.
Included in
UR-0227 Compilation of Binary Neurons to OBDDs
https://www.kennesaw.edu/ccse/events/computing-showcase/fa25-cday-program.php
A neuron with binary inputs & outputs corresponds to a Boolean function. To explain and verify the behavior a neuron (and by extension, a neural network), we can explain and verify its Boolean function. There has been recent interest in representing the Boolean function of such a neuron as an Ordered Binary Decision Diagram (OBDD), which facilitates such analyses. We propose an algorithm for compiling a binary neuron into an OBDD using a compiler that decomposes a Boolean function into a decision graph. We augment this compiler so that it outputs an OBDD instead. Our augmented compiler produces intermediate OBDDs that represent inner- and outer-bounds of the original neuron, tightening compilation progresses. Theoretically, decision graphs of binary neurons are more succinct than their decision trees. Empirically, compilation to decision graphs can scale to neurons with over a thousand features, compared to dozens of features using other compilers.