Understanding Feature Transfer Through Representation Alignment

• URL: http://arxiv.org/abs/2112.07806v1
• Date: Wed, 15 Dec 2021 00:20:29 GMT
• Title: Understanding Feature Transfer Through Representation Alignment
• Authors: Ehsan Imani, Wei Hu, Martha White
• Abstract summary: We find that training neural networks with different architectures and generalizations on random or true labels enforces the same relationship between the hidden representations and the training labels. We show in a classic synthetic transfer problem that alignment is the determining factor for positive and negative transfer to similar and dissimilar tasks.
• Score: 45.35473578109525
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Training with the true labels of a dataset as opposed to randomized labels leads to faster optimization and better generalization. This difference is attributed to a notion of alignment between inputs and labels in natural datasets. We find that training neural networks with different architectures and optimizers on random or true labels enforces the same relationship between the hidden representations and the training labels, elucidating why neural network representations have been so successful for transfer. We first highlight why aligned features promote transfer and show in a classic synthetic transfer problem that alignment is the determining factor for positive and negative transfer to similar and dissimilar tasks. We then investigate a variety of neural network architectures and find that (a) alignment emerges across a variety of different architectures and optimizers, with more alignment arising from depth (b) alignment increases for layers closer to the output and (c) existing high-performance deep CNNs exhibit high levels of alignment.

Related papers

• On Characterizing the Evolution of Embedding Space of Neural Networks using Algebraic Topology [9.537910170141467]
  We study how the topology of feature embedding space changes as it passes through the layers of a well-trained deep neural network (DNN) through Betti numbers. We demonstrate that as depth increases, a topologically complicated dataset is transformed into a simple one, resulting in Betti numbers attaining their lowest possible value.
  arXiv  Detail & Related papers  (2023-11-08T10:45:12Z)
• Network Alignment with Transferable Graph Autoencoders [79.89704126746204]
  We propose a novel graph autoencoder architecture designed to extract powerful and robust node embeddings. We prove that the generated embeddings are associated with the eigenvalues and eigenvectors of the graphs. Our proposed framework also leverages transfer learning and data augmentation to achieve efficient network alignment at a very large scale without retraining.
  arXiv  Detail & Related papers  (2023-10-05T02:58:29Z)
• Domain-adaptive Message Passing Graph Neural Network [67.35534058138387]
  Cross-network node classification (CNNC) aims to classify nodes in a label-deficient target network by transferring the knowledge from a source network with abundant labels. We propose a domain-adaptive message passing graph neural network (DM-GNN), which integrates graph neural network (GNN) with conditional adversarial domain adaptation.
  arXiv  Detail & Related papers  (2023-08-31T05:26:08Z)
• ReduNet: A White-box Deep Network from the Principle of Maximizing Rate Reduction [32.489371527159236]
  This work attempts to provide a plausible theoretical framework that aims to interpret modern deep (convolutional) networks from the principles of data compression and discriminative representation. We show that for high-dimensional multi-class data, the optimal linear discriminative representation maximizes the coding rate difference between the whole dataset and the average of all the subsets. We show that the basic iterative gradient ascent scheme for optimizing the rate reduction objective naturally leads to a multi-layer deep network, named ReduNet, that shares common characteristics of modern deep networks.
  arXiv  Detail & Related papers  (2021-05-21T16:29:57Z)
• Dual-constrained Deep Semi-Supervised Coupled Factorization Network with Enriched Prior [80.5637175255349]
  We propose a new enriched prior based Dual-constrained Deep Semi-Supervised Coupled Factorization Network, called DS2CF-Net. To ex-tract hidden deep features, DS2CF-Net is modeled as a deep-structure and geometrical structure-constrained neural network. Our network can obtain state-of-the-art performance for representation learning and clustering.
  arXiv  Detail & Related papers  (2020-09-08T13:10:21Z)
• Self-Challenging Improves Cross-Domain Generalization [81.99554996975372]
  Convolutional Neural Networks (CNN) conduct image classification by activating dominant features that correlated with labels. We introduce a simple training, Self-Challenging Representation (RSC), that significantly improves the generalization of CNN to the out-of-domain data. RSC iteratively challenges the dominant features activated on the training data, and forces the network to activate remaining features that correlates with labels.
  arXiv  Detail & Related papers  (2020-07-05T21:42:26Z)
• What Do Neural Networks Learn When Trained With Random Labels? [20.54410239839646]
  We study deep neural networks (DNNs) trained on natural image data with entirely random labels. We show analytically for convolutional and fully connected networks that an alignment between the principal components of network parameters and data takes place when training with random labels. We show how this alignment produces a positive transfer: networks pre-trained with random labels train faster downstream compared to training from scratch.
  arXiv  Detail & Related papers  (2020-06-18T12:07:22Z)
• Dynamic Hierarchical Mimicking Towards Consistent Optimization Objectives [73.15276998621582]
  We propose a generic feature learning mechanism to advance CNN training with enhanced generalization ability. Partially inspired by DSN, we fork delicately designed side branches from the intermediate layers of a given neural network. Experiments on both category and instance recognition tasks demonstrate the substantial improvements of our proposed method.
  arXiv  Detail & Related papers  (2020-03-24T09:56:13Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.