Generalization properties of contrastive world models

• URL: http://arxiv.org/abs/2401.00057v1
• Date: Fri, 29 Dec 2023 19:25:34 GMT
• Title: Generalization properties of contrastive world models
• Authors: Kandan Ramakrishnan, R. James Cotton, Xaq Pitkow, Andreas S. Tolias
• Abstract summary: We conduct an extensive study on the generalization properties of contrastive world model. Our experiments show that the contrastive world model fails to generalize under the different OOD tests. Our work highlights the importance of object-centric representations for generalization and current models are limited in their capacity to learn such representations required for human-level generalization.
• Score: 10.806958747213976
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent work on object-centric world models aim to factorize representations in terms of objects in a completely unsupervised or self-supervised manner. Such world models are hypothesized to be a key component to address the generalization problem. While self-supervision has shown improved performance however, OOD generalization has not been systematically and explicitly tested. In this paper, we conduct an extensive study on the generalization properties of contrastive world model. We systematically test the model under a number of different OOD generalization scenarios such as extrapolation to new object attributes, introducing new conjunctions or new attributes. Our experiments show that the contrastive world model fails to generalize under the different OOD tests and the drop in performance depends on the extent to which the samples are OOD. When visualizing the transition updates and convolutional feature maps, we observe that any changes in object attributes (such as previously unseen colors, shapes, or conjunctions of color and shape) breaks down the factorization of object representations. Overall, our work highlights the importance of object-centric representations for generalization and current models are limited in their capacity to learn such representations required for human-level generalization.

Related papers

• In Search of Forgotten Domain Generalization [20.26519807919284]
  Out-of-Domain (OOD) generalization is the ability of a model trained on one or more domains to generalize to unseen domains. In the ImageNet era of computer vision, evaluation sets for measuring a model's OOD performance were designed to be strictly OOD with respect to style. The emergence of foundation models and expansive web-scale datasets has obfuscated this evaluation process.
  arXiv  Detail & Related papers  (2024-10-10T17:50:45Z)
• Unveiling the Generalization Power of Fine-Tuned Large Language Models [81.70754292058258]
  We investigate whether fine-tuning affects the intrinsic generalization ability intrinsic to Large Language Models (LLMs) Our main findings reveal that models fine-tuned on generation and classification tasks exhibit dissimilar behaviors in generalizing to different domains and tasks. We observe that integrating the in-context learning strategy during fine-tuning on generation tasks can enhance the model's generalization ability.
  arXiv  Detail & Related papers  (2024-03-14T08:18:59Z)
• Adaptive Contextual Perception: How to Generalize to New Backgrounds and Ambiguous Objects [75.15563723169234]
  We investigate how vision models adaptively use context for out-of-distribution generalization. We show that models that excel in one setting tend to struggle in the other. To replicate the generalization abilities of biological vision, computer vision models must have factorized object vs. background representations.
  arXiv  Detail & Related papers  (2023-06-09T15:29:54Z)
• On the Compositional Generalization Gap of In-Context Learning [73.09193595292233]
  We look at the gap between the in-distribution (ID) and out-of-distribution (OOD) performance of such models in semantic parsing tasks with in-context learning. We evaluate four model families, OPT, BLOOM, CodeGen and Codex on three semantic parsing datasets.
  arXiv  Detail & Related papers  (2022-11-15T19:56:37Z)
• Translational Concept Embedding for Generalized Compositional Zero-shot Learning [73.60639796305415]
  Generalized compositional zero-shot learning means to learn composed concepts of attribute-object pairs in a zero-shot fashion. This paper introduces a new approach, termed translational concept embedding, to solve these two difficulties in a unified framework.
  arXiv  Detail & Related papers  (2021-12-20T21:27:51Z)
• General Greedy De-bias Learning [163.65789778416172]
  We propose a General Greedy De-bias learning framework (GGD), which greedily trains the biased models and the base model like gradient descent in functional space. GGD can learn a more robust base model under the settings of both task-specific biased models with prior knowledge and self-ensemble biased model without prior knowledge.
  arXiv  Detail & Related papers  (2021-12-20T14:47:32Z)
• Visual Representation Learning Does Not Generalize Strongly Within the Same Domain [41.66817277929783]
  We test whether 17 unsupervised, weakly supervised, and fully supervised representation learning approaches correctly infer the generative factors of variation in simple datasets. We train and test 2000+ models and observe that all of them struggle to learn the underlying mechanism regardless of supervision signal and architectural bias.
  arXiv  Detail & Related papers  (2021-07-17T11:24:18Z)
• Generalization and Robustness Implications in Object-Centric Learning [23.021791024676986]
  In this paper, we train state-of-the-art unsupervised models on five common multi-object datasets. From our experimental study, we find object-centric representations to be generally useful for downstream tasks.
  arXiv  Detail & Related papers  (2021-07-01T17:51:11Z)
• Rethinking Generalization of Neural Models: A Named Entity Recognition Case Study [81.11161697133095]
  We take the NER task as a testbed to analyze the generalization behavior of existing models from different perspectives. Experiments with in-depth analyses diagnose the bottleneck of existing neural NER models. As a by-product of this paper, we have open-sourced a project that involves a comprehensive summary of recent NER papers.
  arXiv  Detail & Related papers  (2020-01-12T04:33:53Z)

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.