Questioning Representational Optimism in Deep Learning: The Fractured Entangled Representation Hypothesis

• URL: http://arxiv.org/abs/2505.11581v1
• Date: Fri, 16 May 2025 16:28:34 GMT
• Title: Questioning Representational Optimism in Deep Learning: The Fractured Entangled Representation Hypothesis
• Authors: Akarsh Kumar, Jeff Clune, Joel Lehman, Kenneth O. Stanley,
• Abstract summary: We compare neural networks evolved through an open-ended search process to networks trained via conventional gradient descent.<n>While both networks produce the same output behavior, their internal representations differ dramatically.<n>In large models, FER may be degrading core model capacities like generalization, creativity, and (continual) learning.
• Score: 14.275283048655268
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Much of the excitement in modern AI is driven by the observation that scaling up existing systems leads to better performance. But does better performance necessarily imply better internal representations? While the representational optimist assumes it must, this position paper challenges that view. We compare neural networks evolved through an open-ended search process to networks trained via conventional stochastic gradient descent (SGD) on the simple task of generating a single image. This minimal setup offers a unique advantage: each hidden neuron's full functional behavior can be easily visualized as an image, thus revealing how the network's output behavior is internally constructed neuron by neuron. The result is striking: while both networks produce the same output behavior, their internal representations differ dramatically. The SGD-trained networks exhibit a form of disorganization that we term fractured entangled representation (FER). Interestingly, the evolved networks largely lack FER, even approaching a unified factored representation (UFR). In large models, FER may be degrading core model capacities like generalization, creativity, and (continual) learning. Therefore, understanding and mitigating FER could be critical to the future of representation learning.

Related papers

• Algorithm Development in Neural Networks: Insights from the Streaming Parity Task [8.188549368578704]
  We study the learning dynamics of neural networks trained on a streaming parity task.<n>We show that, with sufficient finite training experience, RNNs exhibit a phase transition to perfect infinite generalization.<n>Our results disclose one mechanism by which neural networks can generalize infinitely from finite training experience.
  arXiv  Detail & Related papers  (2025-07-14T04:07:43Z)
• Coding schemes in neural networks learning classification tasks [52.22978725954347]
  We investigate fully-connected, wide neural networks learning classification tasks. We show that the networks acquire strong, data-dependent features. Surprisingly, the nature of the internal representations depends crucially on the neuronal nonlinearity.
  arXiv  Detail & Related papers  (2024-06-24T14:50:05Z)
• Unsupervised representation learning with Hebbian synaptic and structural plasticity in brain-like feedforward neural networks [0.0]
  We introduce and evaluate a brain-like neural network model capable of unsupervised representation learning.<n>The model was tested on a diverse set of popular machine learning benchmarks.
  arXiv  Detail & Related papers  (2024-06-07T08:32:30Z)
• Neural networks trained with SGD learn distributions of increasing complexity [78.30235086565388]
  We show that neural networks trained using gradient descent initially classify their inputs using lower-order input statistics. We then exploit higher-order statistics only later during training. We discuss the relation of DSB to other simplicity biases and consider its implications for the principle of universality in learning.
  arXiv  Detail & Related papers  (2022-11-21T15:27:22Z)
• Interpretable part-whole hierarchies and conceptual-semantic relationships in neural networks [4.153804257347222]
  We present Agglomerator, a framework capable of providing a representation of part-whole hierarchies from visual cues. We evaluate our method on common datasets, such as SmallNORB, MNIST, FashionMNIST, CIFAR-10, and CIFAR-100.
  arXiv  Detail & Related papers  (2022-03-07T10:56:13Z)
• Dynamic Inference with Neural Interpreters [72.90231306252007]
  We present Neural Interpreters, an architecture that factorizes inference in a self-attention network as a system of modules. inputs to the model are routed through a sequence of functions in a way that is end-to-end learned. We show that Neural Interpreters perform on par with the vision transformer using fewer parameters, while being transferrable to a new task in a sample efficient manner.
  arXiv  Detail & Related papers  (2021-10-12T23:22:45Z)
• Mitigating Performance Saturation in Neural Marked Point Processes: Architectures and Loss Functions [50.674773358075015]
  We propose a simple graph-based network structure called GCHP, which utilizes only graph convolutional layers. We show that GCHP can significantly reduce training time and the likelihood ratio loss with interarrival time probability assumptions can greatly improve the model performance.
  arXiv  Detail & Related papers  (2021-07-07T16:59:14Z)
• It's FLAN time! Summing feature-wise latent representations for interpretability [0.0]
  We propose a novel class of structurally-constrained neural networks, which we call FLANs (Feature-wise Latent Additive Networks) FLANs process each input feature separately, computing for each of them a representation in a common latent space. These feature-wise latent representations are then simply summed, and the aggregated representation is used for prediction.
  arXiv  Detail & Related papers  (2021-06-18T12:19:33Z)
• Redundant representations help generalization in wide neural networks [71.38860635025907]
  We study the last hidden layer representations of various state-of-the-art convolutional neural networks. We find that if the last hidden representation is wide enough, its neurons tend to split into groups that carry identical information, and differ from each other only by statistically independent noise.
  arXiv  Detail & Related papers  (2021-06-07T10:18:54Z)
• Towards Understanding Hierarchical Learning: Benefits of Neural Representations [160.33479656108926]
  In this work, we demonstrate that intermediate neural representations add more flexibility to neural networks. We show that neural representation can achieve improved sample complexities compared with the raw input. Our results characterize when neural representations are beneficial, and may provide a new perspective on why depth is important in deep learning.
  arXiv  Detail & Related papers  (2020-06-24T02:44:54Z)

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.