On the Informativeness of Supervision Signals

• URL: http://arxiv.org/abs/2211.01407v3
• Date: Tue, 4 Jul 2023 15:25:05 GMT
• Title: On the Informativeness of Supervision Signals
• Authors: Ilia Sucholutsky and Ruairidh M. Battleday and Katherine M. Collins and Raja Marjieh and Joshua C. Peterson and Pulkit Singh and Umang Bhatt and Nori Jacoby and Adrian Weller and Thomas L. Griffiths
• Abstract summary: We use information theory to compare how a number of commonly-used supervision signals contribute to representation-learning performance. Our framework provides theoretical justification for using hard labels in the big-data regime, but richer supervision signals for few-shot learning and out-of-distribution generalization.
• Score: 31.418827619510036
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Supervised learning typically focuses on learning transferable representations from training examples annotated by humans. While rich annotations (like soft labels) carry more information than sparse annotations (like hard labels), they are also more expensive to collect. For example, while hard labels only provide information about the closest class an object belongs to (e.g., "this is a dog"), soft labels provide information about the object's relationship with multiple classes (e.g., "this is most likely a dog, but it could also be a wolf or a coyote"). We use information theory to compare how a number of commonly-used supervision signals contribute to representation-learning performance, as well as how their capacity is affected by factors such as the number of labels, classes, dimensions, and noise. Our framework provides theoretical justification for using hard labels in the big-data regime, but richer supervision signals for few-shot learning and out-of-distribution generalization. We validate these results empirically in a series of experiments with over 1 million crowdsourced image annotations and conduct a cost-benefit analysis to establish a tradeoff curve that enables users to optimize the cost of supervising representation learning on their own datasets.

Related papers

• How many labelers do you have? A closer look at gold-standard labels [10.637125300701795]
  We show how access to non-aggregated label information can make training well-calibrated models more feasible than it is with gold-standard labels. We make several predictions for real-world datasets, including when non-aggregate labels should improve learning performance.
  arXiv  Detail & Related papers  (2022-06-24T02:33:50Z)
• Active Learning with Label Comparisons [41.82179028046654]
  We show that finding the best of $k$ labels can be done with $k-1$ active queries. Key element in our analysis is the "label neighborhood graph" of the true distribution.
  arXiv  Detail & Related papers  (2022-04-10T12:13:46Z)
• Investigating Power laws in Deep Representation Learning [4.996066540156903]
  We propose a framework to evaluate the quality of representations in unlabelled datasets. We estimate the coefficient of the power law, $alpha$, across three key attributes which influence representation learning. Notably, $alpha$ is computable from the representations without knowledge of any labels, thereby offering a framework to evaluate the quality of representations in unlabelled datasets.
  arXiv  Detail & Related papers  (2022-02-11T18:11:32Z)
• Data Consistency for Weakly Supervised Learning [15.365232702938677]
  Training machine learning models involves using large amounts of human-annotated data. We propose a novel weak supervision algorithm that processes noisy labels, i.e., weak signals. We show that it significantly outperforms state-of-the-art weak supervision methods on both text and image classification tasks.
  arXiv  Detail & Related papers  (2022-02-08T16:48:19Z)
• Improving Contrastive Learning on Imbalanced Seed Data via Open-World Sampling [96.8742582581744]
  We present an open-world unlabeled data sampling framework called Model-Aware K-center (MAK) MAK follows three simple principles: tailness, proximity, and diversity. We demonstrate that MAK can consistently improve both the overall representation quality and the class balancedness of the learned features.
  arXiv  Detail & Related papers  (2021-11-01T15:09:41Z)
• Self-supervised Learning is More Robust to Dataset Imbalance [65.84339596595383]
  We investigate self-supervised learning under dataset imbalance. Off-the-shelf self-supervised representations are already more robust to class imbalance than supervised representations. We devise a re-weighted regularization technique that consistently improves the SSL representation quality on imbalanced datasets.
  arXiv  Detail & Related papers  (2021-10-11T06:29:56Z)
• Self-Supervised Learning for Fine-Grained Image Classification [0.0]
  Fine-grained datasets usually provide bounding box annotations along with class labels to aid the process of classification. On the other hand, self-supervised learning exploits the freely available data to generate supervisory signals which act as labels. Our idea is to leverage self-supervision such that the model learns useful representations of fine-grained image classes.
  arXiv  Detail & Related papers  (2021-07-29T14:01:31Z)
• Towards Good Practices for Efficiently Annotating Large-Scale Image Classification Datasets [90.61266099147053]
  We investigate efficient annotation strategies for collecting multi-class classification labels for a large collection of images. We propose modifications and best practices aimed at minimizing human labeling effort. Simulated experiments on a 125k image subset of the ImageNet100 show that it can be annotated to 80% top-1 accuracy with 0.35 annotations per image on average.
  arXiv  Detail & Related papers  (2021-04-26T16:29:32Z)
• Noisy Labels Can Induce Good Representations [53.47668632785373]
  We study how architecture affects learning with noisy labels. We show that training with noisy labels can induce useful hidden representations, even when the model generalizes poorly. This finding leads to a simple method to improve models trained on noisy labels.
  arXiv  Detail & Related papers  (2020-12-23T18:58:05Z)
• Are Fewer Labels Possible for Few-shot Learning? [81.89996465197392]
  Few-shot learning is challenging due to its very limited data and labels. Recent studies in big transfer (BiT) show that few-shot learning can greatly benefit from pretraining on large scale labeled dataset in a different domain. We propose eigen-finetuning to enable fewer shot learning by leveraging the co-evolution of clustering and eigen-samples in the finetuning.
  arXiv  Detail & Related papers  (2020-12-10T18:59:29Z)
• Laplacian Denoising Autoencoder [114.21219514831343]
  We propose to learn data representations with a novel type of denoising autoencoder. The noisy input data is generated by corrupting latent clean data in the gradient domain. Experiments on several visual benchmarks demonstrate that better representations can be learned with the proposed approach.
  arXiv  Detail & Related papers  (2020-03-30T16:52:39Z)

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.