Compositional Generalization Requires More Than Disentangled Representations

• URL: http://arxiv.org/abs/2501.18797v1
• Date: Thu, 30 Jan 2025 23:20:41 GMT
• Title: Compositional Generalization Requires More Than Disentangled Representations
• Authors: Qiyao Liang, Daoyuan Qian, Liu Ziyin, Ila Fiete,
• Abstract summary: compositional generalization remains a key challenge for deep learning. Many generative models fail to recognize and compose factors to generate out-of-distribution (OOD) samples. We show that models forced-through architectural modifications with regularization or curated training data-can be highly data-efficient and effective at learning to compose in OOD regions.
• Score: 5.762286612061953
• License:
• Abstract: Composition-the ability to generate myriad variations from finite means-is believed to underlie powerful generalization. However, compositional generalization remains a key challenge for deep learning. A widely held assumption is that learning disentangled (factorized) representations naturally supports this kind of extrapolation. Yet, empirical results are mixed, with many generative models failing to recognize and compose factors to generate out-of-distribution (OOD) samples. In this work, we investigate a controlled 2D Gaussian "bump" generation task, demonstrating that standard generative architectures fail in OOD regions when training with partial data, even when supplied with fully disentangled $(x, y)$ coordinates, re-entangling them through subsequent layers. By examining the model's learned kernels and manifold geometry, we show that this failure reflects a "memorization" strategy for generation through the superposition of training data rather than by combining the true factorized features. We show that models forced-through architectural modifications with regularization or curated training data-to create disentangled representations in the full-dimensional representational (pixel) space can be highly data-efficient and effective at learning to compose in OOD regions. These findings underscore that bottlenecks with factorized/disentangled representations in an abstract representation are insufficient: the model must actively maintain or induce factorization directly in the representational space in order to achieve robust compositional generalization.

Related papers

• Graph-based Unsupervised Disentangled Representation Learning via Multimodal Large Language Models [42.17166746027585]
  We introduce a bidirectional weighted graph-based framework to learn factorized attributes and their interrelations within complex data. Specifically, we propose a $beta$-VAE based module to extract factors as the initial nodes of the graph. By integrating these complementary modules, our model successfully achieves fine-grained, practical and unsupervised disentanglement.
  arXiv  Detail & Related papers  (2024-07-26T15:32:21Z)
• Learning Divergence Fields for Shift-Robust Graph Representations [73.11818515795761]
  In this work, we propose a geometric diffusion model with learnable divergence fields for the challenging problem with interdependent data. We derive a new learning objective through causal inference, which can guide the model to learn generalizable patterns of interdependence that are insensitive across domains.
  arXiv  Detail & Related papers  (2024-06-07T14:29:21Z)
• Compositional Abilities Emerge Multiplicatively: Exploring Diffusion Models on a Synthetic Task [20.749514363389878]
  We study compositional generalization in conditional diffusion models in a synthetic setting. We find that the order in which the ability to generate samples emerges is governed by the structure of the underlying data-generating process. Our study lays a foundation for understanding capabilities and compositionality in generative models from a data-centric perspective.
  arXiv  Detail & Related papers  (2023-10-13T18:00:59Z)
• Flow Factorized Representation Learning [109.51947536586677]
  We introduce a generative model which specifies a distinct set of latent probability paths that define different input transformations. We show that our model achieves higher likelihoods on standard representation learning benchmarks while simultaneously being closer to approximately equivariant models.
  arXiv  Detail & Related papers  (2023-09-22T20:15:37Z)
• Representation Disentaglement via Regularization by Causal Identification [3.9160947065896803]
  We propose the use of a causal collider structured model to describe the underlying data generative process assumptions in disentangled representation learning. For this, we propose regularization by identification (ReI), a modular regularization engine designed to align the behavior of large scale generative models with the disentanglement constraints imposed by causal identification.
  arXiv  Detail & Related papers  (2023-02-28T23:18:54Z)
• DIFFormer: Scalable (Graph) Transformers Induced by Energy Constrained Diffusion [66.21290235237808]
  We introduce an energy constrained diffusion model which encodes a batch of instances from a dataset into evolutionary states. We provide rigorous theory that implies closed-form optimal estimates for the pairwise diffusion strength among arbitrary instance pairs. Experiments highlight the wide applicability of our model as a general-purpose encoder backbone with superior performance in various tasks.
  arXiv  Detail & Related papers  (2023-01-23T15:18:54Z)
• Dynamic Latent Separation for Deep Learning [67.62190501599176]
  A core problem in machine learning is to learn expressive latent variables for model prediction on complex data. Here, we develop an approach that improves expressiveness, provides partial interpretation, and is not restricted to specific applications.
  arXiv  Detail & Related papers  (2022-10-07T17:56:53Z)
• Towards Understanding and Mitigating Dimensional Collapse in Heterogeneous Federated Learning [112.69497636932955]
  Federated learning aims to train models across different clients without the sharing of data for privacy considerations. We study how data heterogeneity affects the representations of the globally aggregated models. We propose sc FedDecorr, a novel method that can effectively mitigate dimensional collapse in federated learning.
  arXiv  Detail & Related papers  (2022-10-01T09:04:17Z)
• Global and Local Features through Gaussian Mixture Models on Image Semantic Segmentation [0.38073142980732994]
  We propose an internal structure for the feature representations while extracting a global representation that supports the former. During training, we predict a Gaussian Mixture Model from the data, which, merged with the skip connections and the decoding stage, helps avoid wrong inductive biases. Our results show that we can improve semantic segmentation by providing both learning representations (global and local) with a clustering behavior and combining them.
  arXiv  Detail & Related papers  (2022-07-19T10:10:49Z)
• Amortized Inference for Causal Structure Learning [72.84105256353801]
  Learning causal structure poses a search problem that typically involves evaluating structures using a score or independence test. We train a variational inference model to predict the causal structure from observational/interventional data. Our models exhibit robust generalization capabilities under substantial distribution shift.
  arXiv  Detail & Related papers  (2022-05-25T17:37:08Z)
• Hierarchical regularization networks for sparsification based learning on noisy datasets [0.0]
  hierarchy follows from approximation spaces identified at successively finer scales. For promoting model generalization at each scale, we also introduce a novel, projection based penalty operator across multiple dimension. Results show the performance of the approach as a data reduction and modeling strategy on both synthetic and real datasets.
  arXiv  Detail & Related papers  (2020-06-09T18:32:24Z)

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.