Unraveling the Enigma of Double Descent: An In-depth Analysis through the Lens of Learned Feature Space

• URL: http://arxiv.org/abs/2310.13572v3
• Date: Thu, 25 Apr 2024 10:37:00 GMT
• Title: Unraveling the Enigma of Double Descent: An In-depth Analysis through the Lens of Learned Feature Space
• Authors: Yufei Gu, Xiaoqing Zheng, Tomaso Aste,
• Abstract summary: Double descent presents a counter-intuitive aspect within the machine learning domain. We argue that double descent arises in imperfect models trained with noisy data.
• Score: 12.907949196758565
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Double descent presents a counter-intuitive aspect within the machine learning domain, and researchers have observed its manifestation in various models and tasks. While some theoretical explanations have been proposed for this phenomenon in specific contexts, an accepted theory to account for its occurrence in deep learning remains yet to be established. In this study, we revisit the phenomenon of double descent and demonstrate that its occurrence is strongly influenced by the presence of noisy data. Through conducting a comprehensive analysis of the feature space of learned representations, we unveil that double descent arises in imperfect models trained with noisy data. We argue that double descent is a consequence of the model first learning the noisy data until interpolation and then adding implicit regularization via over-parameterization acquiring therefore capability to separate the information from the noise.

Related papers

• Multiple Descents in Unsupervised Learning: The Role of Noise, Domain Shift and Anomalies [14.399035468023161]
  We study the presence of double descent in unsupervised learning, an area that has received little attention and is not yet fully understood. We use synthetic and real data and identify model-wise, epoch-wise, and sample-wise double descent for various applications.
  arXiv  Detail & Related papers  (2024-06-17T16:24:23Z)
• Class-wise Activation Unravelling the Engima of Deep Double Descent [0.0]
  Double descent presents a counter-intuitive aspect within the machine learning domain. In this study, we revisited the phenomenon of double descent and discussed the conditions of its occurrence.
  arXiv  Detail & Related papers  (2024-05-13T12:07:48Z)
• Unified View of Grokking, Double Descent and Emergent Abilities: A Perspective from Circuits Competition [83.13280812128411]
  Recent studies have uncovered intriguing phenomena in deep learning, such as grokking, double descent, and emergent abilities in large language models. We present a comprehensive framework that provides a unified view of these three phenomena, focusing on the competition between memorization and generalization circuits.
  arXiv  Detail & Related papers  (2024-02-23T08:14:36Z)
• Neural Causal Abstractions [63.21695740637627]
  We develop a new family of causal abstractions by clustering variables and their domains. We show that such abstractions are learnable in practical settings through Neural Causal Models. Our experiments support the theory and illustrate how to scale causal inferences to high-dimensional settings involving image data.
  arXiv  Detail & Related papers  (2024-01-05T02:00:27Z)
• A U-turn on Double Descent: Rethinking Parameter Counting in Statistical Learning [68.76846801719095]
  We show that double descent appears exactly when and where it occurs, and that its location is not inherently tied to the threshold p=n. This provides a resolution to tensions between double descent and statistical intuition.
  arXiv  Detail & Related papers  (2023-10-29T12:05:39Z)
• Unveiling the Potential of Probabilistic Embeddings in Self-Supervised Learning [4.124934010794795]
  Self-supervised learning has played a pivotal role in advancing machine learning by allowing models to acquire meaningful representations from unlabeled data. We investigate the impact of probabilistic modeling on the information bottleneck, shedding light on a trade-off between compression and preservation of information in both representation and loss space. Our findings suggest that introducing an additional bottleneck in the loss space can significantly enhance the ability to detect out-of-distribution examples.
  arXiv  Detail & Related papers  (2023-10-27T12:01:16Z)
• On the Joint Interaction of Models, Data, and Features [82.60073661644435]
  We introduce a new tool, the interaction tensor, for empirically analyzing the interaction between data and model through features. Based on these observations, we propose a conceptual framework for feature learning. Under this framework, the expected accuracy for a single hypothesis and agreement for a pair of hypotheses can both be derived in closed-form.
  arXiv  Detail & Related papers  (2023-06-07T21:35:26Z)
• Inverse Dynamics Pretraining Learns Good Representations for Multitask Imitation [66.86987509942607]
  We evaluate how such a paradigm should be done in imitation learning. We consider a setting where the pretraining corpus consists of multitask demonstrations. We argue that inverse dynamics modeling is well-suited to this setting.
  arXiv  Detail & Related papers  (2023-05-26T14:40:46Z)
• Towards Causal Representation Learning and Deconfounding from Indefinite Data [17.793702165499298]
  Non-statistical data (e.g., images, text, etc.) encounters significant conflicts in terms of properties and methods with traditional causal data. We redefine causal data from two novel perspectives and then propose three data paradigms. We implement the above designs as a dynamic variational inference model, tailored to learn causal representation from indefinite data.
  arXiv  Detail & Related papers  (2023-05-04T08:20:37Z)
• Multi-scale Feature Learning Dynamics: Insights for Double Descent [71.91871020059857]
  We study the phenomenon of "double descent" of the generalization error. We find that double descent can be attributed to distinct features being learned at different scales.
  arXiv  Detail & Related papers  (2021-12-06T18:17:08Z)

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.