Sequencing the Neurome: Towards Scalable Exact Parameter Reconstruction of Black-Box Neural Networks

TL;DR

This work presents a novel query generation algorithm that produces maximally informative samples, letting us untangle the non-linear relationships efficiently, and demonstrates robustness and scalability across a variety of architectures, datasets, and training procedures.

Abstract

Inferring the exact parameters of a neural network with only query access is an NP-Hard problem, with few practical existing algorithms. Solutions would have major implications for security, verification, interpretability, and understanding biological networks. The key challenges are the massive parameter space, and complex non-linear relationships between neurons. We resolve these challenges using two insights. First, we observe that almost all networks used in practice are produced by random initialization and first order optimization, an inductive bias that drastically reduces the practical parameter space. Second, we present a novel query generation algorithm that produces maximally informative samples, letting us untangle the non-linear relationships efficiently. We demonstrate reconstruction of a hidden network containing over 1.5 million parameters, and of one 7 layers deep, the largest and deepest reconstructions to date, with max parameter difference less than 0.0001, and illustrate robustness and scalability across a variety of architectures, datasets, and training procedures.

Figures (8)

• Figure 1: Problem Overview and Illustration of Reconstruction Algorithm and Query Generation Algorithm
• Figure 2: Illustration of Network Isomorphisms
• Figure 3: Illustration of Performance Convergence for 784x128x80x40x32x16x10 Network
• Figure 4: Illustration of Performance Convergence and Alignment Stability
• Figure 5: Illustration of Architectures, and their properties of Convergence and Divergence