MoDE: CLIP Data Experts via Clustering

• URL: http://arxiv.org/abs/2404.16030v1
• Date: Wed, 24 Apr 2024 17:59:24 GMT
• Title: MoDE: CLIP Data Experts via Clustering
• Authors: Jiawei Ma, Po-Yao Huang, Saining Xie, Shang-Wen Li, Luke Zettlemoyer, Shih-Fu Chang, Wen-Tau Yih, Hu Xu,
• Abstract summary: We present Mixture of Data Experts (MoDE) and learn a system of CLIP data experts via clustering. Each data expert is trained on one data cluster, being less sensitive to false negative noises in other clusters. Experiments show that four CLIP data experts on ViT-B/16 outperform the ViT-L/14 by OpenAI CLIP and OpenCLIP on zero-shot image classification.
• Score: 119.88225963429569
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: The success of contrastive language-image pretraining (CLIP) relies on the supervision from the pairing between images and captions, which tends to be noisy in web-crawled data. We present Mixture of Data Experts (MoDE) and learn a system of CLIP data experts via clustering. Each data expert is trained on one data cluster, being less sensitive to false negative noises in other clusters. At inference time, we ensemble their outputs by applying weights determined through the correlation between task metadata and cluster conditions. To estimate the correlation precisely, the samples in one cluster should be semantically similar, but the number of data experts should still be reasonable for training and inference. As such, we consider the ontology in human language and propose to use fine-grained cluster centers to represent each data expert at a coarse-grained level. Experimental studies show that four CLIP data experts on ViT-B/16 outperform the ViT-L/14 by OpenAI CLIP and OpenCLIP on zero-shot image classification but with less ($<$35\%) training cost. Meanwhile, MoDE can train all data expert asynchronously and can flexibly include new data experts. The code is available at https://github.com/facebookresearch/MetaCLIP/tree/main/mode.

Related papers

• FedClust: Tackling Data Heterogeneity in Federated Learning through Weight-Driven Client Clustering [26.478852701376294]
  Federated learning (FL) is an emerging distributed machine learning paradigm. One of the major challenges in FL is the presence of uneven data distributions across client devices. We propose em FedClust, a novel approach for CFL that leverages the correlation between local model weights and the data distribution of clients.
  arXiv  Detail & Related papers  (2024-07-09T02:47:16Z)
• Task-customized Masked AutoEncoder via Mixture of Cluster-conditional Experts [104.9871176044644]
  Masked Autoencoder(MAE) is a prevailing self-supervised learning method that achieves promising results in model pre-training. We propose a novel MAE-based pre-training paradigm, Mixture of Cluster-conditional Experts (MoCE) MoCE trains each expert only with semantically relevant images by using cluster-conditional gates.
  arXiv  Detail & Related papers  (2024-02-08T03:46:32Z)
• Demystifying CLIP Data [86.34045746910114]
  Contrastive Language-Image Pre-training (CLIP) has advanced research and applications in computer vision. We introduce Metadata-Curated Language-Image Pre-training (MetaCLIP) MetaCLIP takes a raw data pool and metadata (derived from CLIP's concepts) and yields a balanced subset over the metadata distribution.
  arXiv  Detail & Related papers  (2023-09-28T17:59:56Z)
• CLC: Cluster Assignment via Contrastive Representation Learning [9.631532215759256]
  We propose Contrastive Learning-based Clustering (CLC), which uses contrastive learning to directly learn cluster assignment. We achieve 53.4% accuracy on the full ImageNet dataset and outperform existing methods by large margins.
  arXiv  Detail & Related papers  (2023-06-08T07:15:13Z)
• ClusterNet: A Perception-Based Clustering Model for Scattered Data [16.326062082938215]
  Cluster separation in scatterplots is a task that is typically tackled by widely used clustering techniques. We propose a learning strategy which directly operates on scattered data. We train ClusterNet, a point-based deep learning model, trained to reflect human perception of cluster separability.
  arXiv  Detail & Related papers  (2023-04-27T13:41:12Z)
• DataComp: In search of the next generation of multimodal datasets [179.79323076587255]
  DataComp is a testbed for dataset experiments centered around a new candidate pool of 12.8 billion image-text pairs from Common Crawl. Our benchmark consists of multiple compute scales spanning four orders of magnitude. In particular, our best baseline, DataComp-1B, enables training a CLIP ViT-L/14 from scratch to 79.2% zero-shot accuracy on ImageNet.
  arXiv  Detail & Related papers  (2023-04-27T11:37:18Z)
• Hard Regularization to Prevent Deep Online Clustering Collapse without Data Augmentation [65.268245109828]
  Online deep clustering refers to the joint use of a feature extraction network and a clustering model to assign cluster labels to each new data point or batch as it is processed. While faster and more versatile than offline methods, online clustering can easily reach the collapsed solution where the encoder maps all inputs to the same point and all are put into a single cluster. We propose a method that does not require data augmentation, and that, differently from existing methods, regularizes the hard assignments.
  arXiv  Detail & Related papers  (2023-03-29T08:23:26Z)
• CADIS: Handling Cluster-skewed Non-IID Data in Federated Learning with Clustered Aggregation and Knowledge DIStilled Regularization [3.3711670942444014]
  Federated learning enables edge devices to train a global model collaboratively without exposing their data. We tackle a new type of Non-IID data, called cluster-skewed non-IID, discovered in actual data sets. We propose an aggregation scheme that guarantees equality between clusters.
  arXiv  Detail & Related papers  (2023-02-21T02:53:37Z)
• You Never Cluster Alone [150.94921340034688]
  We extend the mainstream contrastive learning paradigm to a cluster-level scheme, where all the data subjected to the same cluster contribute to a unified representation. We define a set of categorical variables as clustering assignment confidence, which links the instance-level learning track with the cluster-level one. By reparametrizing the assignment variables, TCC is trained end-to-end, requiring no alternating steps.
  arXiv  Detail & Related papers  (2021-06-03T14:59:59Z)
• Meta Clustering for Collaborative Learning [13.003650251457193]
  In collaborative learning, learners coordinate to enhance each of their learning performances. From the perspective of any learner, a critical challenge is to filter out unqualified collaborators. We propose a framework named meta clustering to address the challenge.
  arXiv  Detail & Related papers  (2020-05-29T21:11:59Z)

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.