Formation of Representations in Neural Networks

• URL: http://arxiv.org/abs/2410.03006v1
• Date: Thu, 3 Oct 2024 21:31:01 GMT
• Title: Formation of Representations in Neural Networks
• Authors: Liu Ziyin, Isaac Chuang, Tomer Galanti, Tomaso Poggio,
• Abstract summary: How complex, structured, and transferable representations emerge in modern neural networks has remained a mystery. We propose the Canonical Representation Hypothesis (CRH), which posits a set of six alignment relations to universally govern the formation of representations. We show that the breaking of CRH leads to the emergence of reciprocal power-law relations between R, W, and G, which we refer to as the Polynomial Alignment Hypothesis (PAH)
• Score: 8.79431718760617
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Understanding neural representations will help open the black box of neural networks and advance our scientific understanding of modern AI systems. However, how complex, structured, and transferable representations emerge in modern neural networks has remained a mystery. Building on previous results, we propose the Canonical Representation Hypothesis (CRH), which posits a set of six alignment relations to universally govern the formation of representations in most hidden layers of a neural network. Under the CRH, the latent representations (R), weights (W), and neuron gradients (G) become mutually aligned during training. This alignment implies that neural networks naturally learn compact representations, where neurons and weights are invariant to task-irrelevant transformations. We then show that the breaking of CRH leads to the emergence of reciprocal power-law relations between R, W, and G, which we refer to as the Polynomial Alignment Hypothesis (PAH). We present a minimal-assumption theory demonstrating that the balance between gradient noise and regularization is crucial for the emergence the canonical representation. The CRH and PAH lead to an exciting possibility of unifying major key deep learning phenomena, including neural collapse and the neural feature ansatz, in a single framework.

Related papers

• Artificial Kuramoto Oscillatory Neurons [65.16453738828672]
  We introduce Artificial Kuramotoy Neurons (AKOrN) as a dynamical alternative to threshold units. We show that this idea provides performance improvements across a wide spectrum of tasks. We believe that these empirical results show the importance of our assumptions at the most basic neuronal level of neural representation.
  arXiv  Detail & Related papers  (2024-10-17T17:47:54Z)
• Collective variables of neural networks: empirical time evolution and scaling laws [0.535514140374842]
  We show that certain measures on the spectrum of the empirical neural tangent kernel, specifically entropy and trace, yield insight into the representations learned by a neural network. Results are demonstrated first on test cases before being shown on more complex networks, including transformers, auto-encoders, graph neural networks, and reinforcement learning studies.
  arXiv  Detail & Related papers  (2024-10-09T21:37:14Z)
• Contrastive Learning in Memristor-based Neuromorphic Systems [55.11642177631929]
  Spiking neural networks have become an important family of neuron-based models that sidestep many of the key limitations facing modern-day backpropagation-trained deep networks. In this work, we design and investigate a proof-of-concept instantiation of contrastive-signal-dependent plasticity (CSDP), a neuromorphic form of forward-forward-based, backpropagation-free learning.
  arXiv  Detail & Related papers  (2024-09-17T04:48:45Z)
• Novel Kernel Models and Exact Representor Theory for Neural Networks Beyond the Over-Parameterized Regime [52.00917519626559]
  This paper presents two models of neural-networks and their training applicable to neural networks of arbitrary width, depth and topology. We also present an exact novel representor theory for layer-wise neural network training with unregularized gradient descent in terms of a local-extrinsic neural kernel (LeNK) This representor theory gives insight into the role of higher-order statistics in neural network training and the effect of kernel evolution in neural-network kernel models.
  arXiv  Detail & Related papers  (2024-05-24T06:30:36Z)
• Gradient Descent in Neural Networks as Sequential Learning in RKBS [63.011641517977644]
  We construct an exact power-series representation of the neural network in a finite neighborhood of the initial weights. We prove that, regardless of width, the training sequence produced by gradient descent can be exactly replicated by regularized sequential learning.
  arXiv  Detail & Related papers  (2023-02-01T03:18:07Z)
• Extrapolation and Spectral Bias of Neural Nets with Hadamard Product: a Polynomial Net Study [55.12108376616355]
  The study on NTK has been devoted to typical neural network architectures, but is incomplete for neural networks with Hadamard products (NNs-Hp) In this work, we derive the finite-width-K formulation for a special class of NNs-Hp, i.e., neural networks. We prove their equivalence to the kernel regression predictor with the associated NTK, which expands the application scope of NTK.
  arXiv  Detail & Related papers  (2022-09-16T06:36:06Z)
• Data-driven emergence of convolutional structure in neural networks [83.4920717252233]
  We show how fully-connected neural networks solving a discrimination task can learn a convolutional structure directly from their inputs. By carefully designing data models, we show that the emergence of this pattern is triggered by the non-Gaussian, higher-order local structure of the inputs.
  arXiv  Detail & Related papers  (2022-02-01T17:11:13Z)
• On 1/n neural representation and robustness [13.491651740693705]
  We show that imposing the experimentally observed structure on artificial neural networks makes them more robust to adversarial attacks. Our findings complement the existing theory relating wide neural networks to kernel methods.
  arXiv  Detail & Related papers  (2020-12-08T20:34:49Z)
• Banach Space Representer Theorems for Neural Networks and Ridge Splines [17.12783792226575]
  We develop a variational framework to understand the properties of the functions learned by neural networks fit to data. We derive a representer theorem showing that finite-width, single-hidden layer neural networks are solutions to inverse problems.
  arXiv  Detail & Related papers  (2020-06-10T02:57:37Z)

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.