A mini-batch training strategy for deep subspace clustering networks

• URL: http://arxiv.org/abs/2507.19917v1
• Date: Sat, 26 Jul 2025 11:44:39 GMT
• Title: A mini-batch training strategy for deep subspace clustering networks
• Authors: Yuxuan Jiang, Chenwei Yu, Zhi Lin, Xiaolan Liu,
• Abstract summary: Mini-batch training is a cornerstone of modern deep learning, offering computational efficiency and scalability for training complex architectures.<n>In this work, we introduce a mini-batch training strategy for deep subspace clustering by integrating a memory bank that preserves global feature representations.<n>Our approach enables scalable training of deep architectures for subspace clustering with high-resolution images, overcoming previous limitations.
• Score: 6.517972913340111
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Mini-batch training is a cornerstone of modern deep learning, offering computational efficiency and scalability for training complex architectures. However, existing deep subspace clustering (DSC) methods, which typically combine an autoencoder with a self-expressive layer, rely on full-batch processing. The bottleneck arises from the self-expressive module, which requires representations of the entire dataset to construct a self-representation coefficient matrix. In this work, we introduce a mini-batch training strategy for DSC by integrating a memory bank that preserves global feature representations. Our approach enables scalable training of deep architectures for subspace clustering with high-resolution images, overcoming previous limitations. Additionally, to efficiently fine-tune large-scale pre-trained encoders for subspace clustering, we propose a decoder-free framework that leverages contrastive learning instead of autoencoding for representation learning. This design not only eliminates the computational overhead of decoder training but also provides competitive performance. Extensive experiments demonstrate that our approach not only achieves performance comparable to full-batch methods, but outperforms other state-of-the-art subspace clustering methods on the COIL100 and ORL datasets by fine-tuning deep networks.

Related papers

• Private Training & Data Generation by Clustering Embeddings [74.00687214400021]
  Differential privacy (DP) provides a robust framework for protecting individual data.<n>We introduce a novel principled method for DP synthetic image embedding generation.<n> Empirically, a simple two-layer neural network trained on synthetically generated embeddings achieves state-of-the-art (SOTA) classification accuracy.
  arXiv  Detail & Related papers  (2025-06-20T00:17:14Z)
• SGLP: A Similarity Guided Fast Layer Partition Pruning for Compressing Large Deep Models [19.479746878680707]
  Layer pruning is a potent approach to reduce network size and improve computational efficiency. We propose a Similarity Guided fast Layer Partition pruning for compressing large deep models. Our method outperforms the state-of-the-art methods in both accuracy and computational efficiency.
  arXiv  Detail & Related papers  (2024-10-14T04:01:08Z)
• Low-Resolution Self-Attention for Semantic Segmentation [93.30597515880079]
  We introduce the Low-Resolution Self-Attention (LRSA) mechanism to capture global context at a significantly reduced computational cost.<n>Our approach involves computing self-attention in a fixed low-resolution space regardless of the input image's resolution.<n>We demonstrate the effectiveness of our LRSA approach by building the LRFormer, a vision transformer with an encoder-decoder structure.
  arXiv  Detail & Related papers  (2023-10-08T06:10:09Z)
• ConvBLS: An Effective and Efficient Incremental Convolutional Broad Learning System for Image Classification [63.49762079000726]
  We propose a convolutional broad learning system (ConvBLS) based on the spherical K-means (SKM) algorithm and two-stage multi-scale (TSMS) feature fusion. Our proposed ConvBLS method is unprecedentedly efficient and effective.
  arXiv  Detail & Related papers  (2023-04-01T04:16:12Z)
• Unifying Synergies between Self-supervised Learning and Dynamic Computation [53.66628188936682]
  We present a novel perspective on the interplay between SSL and DC paradigms. We show that it is feasible to simultaneously learn a dense and gated sub-network from scratch in a SSL setting. The co-evolution during pre-training of both dense and gated encoder offers a good accuracy-efficiency trade-off.
  arXiv  Detail & Related papers  (2023-01-22T17:12:58Z)
• COMET: A Comprehensive Cluster Design Methodology for Distributed Deep Learning Training [42.514897110537596]
  Modern Deep Learning (DL) models have grown to sizes requiring massive clusters of specialized, high-end nodes to train. designing such clusters to maximize both performance and utilization--to amortize their steep cost--is a challenging task. We introduce COMET, a holistic cluster design methodology and workflow to jointly study the impact of parallelization strategies and key cluster resource provisioning on the performance of distributed DL training.
  arXiv  Detail & Related papers  (2022-11-30T00:32:37Z)
• A Deep Dive into Deep Cluster [0.2578242050187029]
  DeepCluster is a simple and scalable unsupervised pretraining of visual representations. We show that DeepCluster convergence and performance depend on the interplay between the quality of the randomly filters of the convolutional layer and the selected number of clusters.
  arXiv  Detail & Related papers  (2022-07-24T22:55:09Z)
• Doing More by Doing Less: How Structured Partial Backpropagation Improves Deep Learning Clusters [9.17259958324486]
  Training deep learning models is resource-intensive, consuming significant compute, memory, and network resources. We propose Structured Partial Backpropagation(SPB), a technique that controls the amount of backpropagation at individual workers in distributed training. We find that JigSaw can improve large scale cluster efficiency by as high as 28%.
  arXiv  Detail & Related papers  (2021-11-20T20:34:26Z)
• DANCE: DAta-Network Co-optimization for Efficient Segmentation Model Training and Inference [86.03382625531951]
  DANCE is an automated simultaneous data-network co-optimization for efficient segmentation model training and inference.<n>It integrates automated data slimming which adaptively downsamples/drops input images and controls their corresponding contribution to the training loss guided by the images' spatial complexity.<n>Experiments and ablating studies demonstrate that DANCE can achieve "all-win" towards efficient segmentation.
  arXiv  Detail & Related papers  (2021-07-16T04:58:58Z)
• Very Compact Clusters with Structural Regularization via Similarity and Connectivity [3.779514860341336]
  We propose an end-to-end deep clustering algorithm, i.e., Very Compact Clusters (VCC) for the general datasets. Our proposed approach achieves better clustering performance over most of the state-of-the-art clustering methods.
  arXiv  Detail & Related papers  (2021-06-09T23:22:03Z)
• Overcomplete Deep Subspace Clustering Networks [80.16644725886968]
  Experimental results on four benchmark datasets show the effectiveness of the proposed method over DSC and other clustering methods in terms of clustering error. Our method is also not as dependent as DSC is on where pre-training should be stopped to get the best performance and is also more robust to noise.
  arXiv  Detail & Related papers  (2020-11-16T22:07:18Z)

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.