Pruned Neural Networks are Surprisingly Modular
The learned weights of a neural network are often considered devoid of scrutable internal structure. To discern structure in these weights, we introduce a measurable notion of modularity for multi-layer perceptrons (MLPs), and investigate the modular structure of MLPs trained on datasets of small images. Our notion of modularity comes from the graph clustering literature: a "module" is a set of neurons with strong internal connectivity but weak external connectivity. We find that training and weight pruning produces MLPs that are more modular than randomly initialized ones, and often significantly more modular than random MLPs with the same (sparse) distribution of weights. Interestingly, they are much more modular when trained with dropout. We also present exploratory analyses of the importance of different modules for performance and how modules depend on each other. Understanding the modular structure of neural networks, when such structure exists, will hopefully render their inner workings more interpretable to engineers. Note that this paper has been superceded by "Clusterability in Neural Networks", arxiv:2103.03386 and "Quantifying Local Specialization in Deep Neural Networks", arxiv:2110.08058!
Code (1)
Tasks
ClusteringGraph ClusteringMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
Scaling Submodular Maximization via Pruned Submodularity Graphs
We propose a new random pruning method (called "submodular sparsification (SS)") to reduce the cost of submodular maximization. The pruning is applied via a "submodularity graph" over the $n$ ground elements, where each …
Video SummarizationScalable Submodular Policy Optimization via Pruned Submodularity Graph
In Reinforcement Learning (abbreviated as RL), an agent interacts with the environment via a set of possible actions, and a reward is generated from some unknown distribution. The task here is to find an optimal set of a…
Reinforcement LearningData-Efficient Structured Pruning via Submodular Optimization
Structured pruning is an effective approach for compressing large pre-trained neural networks without significantly affecting their performance. However, most current structured pruning methods do not provide any perform…
Investigating Hallucinations in Pruned Large Language Models for Abstractive Summarization
Despite the remarkable performance of generative large language models (LLMs) on abstractive summarization, they face two significant challenges: their considerable size and tendency to hallucinate. Hallucinations are co…
Abstractive Text SummarizationHallucinationHallucination EvaluationOn the Interplay Between Sparsity, Naturalness, Intelligibility, and Prosody in Speech Synthesis
Are end-to-end text-to-speech (TTS) models over-parametrized? To what extent can these models be pruned, and what happens to their synthesis capabilities? This work serves as a starting point to explore pruning both spec…
Knowledge DistillationSpeech Synthesistext-to-speechText to Speech