Interpreting the Residual Stream of ResNet18
원문 초록
A mechanistic understanding of the computations learned by deep neural networks (DNNs) is far from complete. In the domain of visual object recognition, prior research has illuminated inner workings of InceptionV1, but DNNs with different architectures have remained largely unexplored. This work investigates ResNet18 with a particular focus on its residual stream, an architectural mechanism which InceptionV1 lacks. We observe that for a given block, channel features of the stream are updated along a spectrum: either the input feature skips to the output, the block feature overwrites the output, or the output is some mixture between the input and block features. Furthermore, we show that many residual stream channels compute scale invariant representations through a mixture of the input's smaller-scale feature with the block's larger-scale feature. This not only mounts evidence for the universality of scale equivariance, but also presents how the residual stream further implements scale invariance. Collectively, our results begin an interpretation of the residual stream in visual object recognition, finding it to be a flexible feature manager and a medium to build scale invariant representations.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
SASep: Saliency-Aware Structured Separation of Geometry and Feature for Open Set Learning on Point Clouds
STSBench: A Spatio-temporal Scenario Benchmark for Multi-modal Large Language Models in Autonomous Driving
InstructSAM: A Training-Free Framework for Instruction-Oriented Remote Sensing Object Recognition
Improving Unsupervised Task-driven Models of Ventral Visual Stream via Relative Position Predictivity
Topology-Guided Knowledge Distillation for Efficient Point Cloud Processing
Naturally Computed Scale Invariance in the Residual Stream of ResNet18
토론
이 논문과 연결된 포럼 글이 아직 없습니다.