Quanshi Zhang

Mining Interpretable AOG Representations from Convolutional Networks via Active Question Answering

2018 · arXiv (Cornell University)

In this paper, we present a method to mine object-part patterns from conv-layers of a pre-trained convolutional neural network (CNN). The mined object-part patterns are organized by an And-Or graph (AOG). This interpretable AOG representation …

Batch Normalization Is Blind to the First and Second Derivatives of the Loss

2022 · arXiv (Cornell University)

In this paper, we prove the effects of the BN operation on the back-propagation of the first and second derivatives of the loss. When we do the Taylor series expansion of the loss function, we …

Can the Inference Logic of Large Language Models be Disentangled into Symbolic Concepts?

2023 · arXiv (Cornell University)

In this paper, we explain the inference logic of large language models (LLMs) as a set of symbolic concepts. Many recent studies have discovered that traditional DNNs usually encode sparse symbolic concepts. However, because an …