Low Rank and Sparse Fourier Structure in Recurrent Networks Trained on Modular Addition

• URL: http://arxiv.org/abs/2503.22059v1
• Date: Fri, 28 Mar 2025 00:40:03 GMT
• Title: Low Rank and Sparse Fourier Structure in Recurrent Networks Trained on Modular Addition
• Authors: Akshay Rangamani,
• Abstract summary: We show that Recurrent Neural Networks (RNNs) trained on modular addition tasks also use a Fourier multiplication strategy.<n>We also show empirically that the RNN is robust to removing individual frequencies, while the performance degrades drastically as more frequencies are ablated from the model.
• Score: 2.973331166114387
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Modular addition tasks serve as a useful test bed for observing empirical phenomena in deep learning, including the phenomenon of \emph{grokking}. Prior work has shown that one-layer transformer architectures learn Fourier Multiplication circuits to solve modular addition tasks. In this paper, we show that Recurrent Neural Networks (RNNs) trained on modular addition tasks also use a Fourier Multiplication strategy. We identify low rank structures in the model weights, and attribute model components to specific Fourier frequencies, resulting in a sparse representation in the Fourier space. We also show empirically that the RNN is robust to removing individual frequencies, while the performance degrades drastically as more frequencies are ablated from the model.

Related papers

• Robustifying Fourier Features Embeddings for Implicit Neural Representations [25.725097757343367]
  Implicit Neural Representations (INRs) employ neural networks to represent continuous functions by mapping coordinates to the corresponding values of the target function.<n>INRs face a challenge known as spectral bias when dealing with scenes containing varying frequencies.<n>We propose the use of multi-layer perceptrons (MLPs) without additive.
  arXiv  Detail & Related papers  (2025-02-08T07:43:37Z)
• Neural Fourier Modelling: A Highly Compact Approach to Time-Series Analysis [9.969451740838418]
  We introduce Neural Fourier Modelling (NFM), a compact yet powerful solution for time-series analysis. NFM is grounded in two key properties of the Fourier transform (FT): (i) the ability to model finite-length time series as functions in the Fourier domain, and (ii) the capacity for data manipulation within the Fourier domain. NFM achieves state-of-the-art performance on a wide range of tasks, including challenging time-series scenarios with previously unseen sampling rates at test time.
  arXiv  Detail & Related papers  (2024-10-07T02:39:55Z)
• Implicit Neural Representations with Fourier Kolmogorov-Arnold Networks [4.499833362998488]
  Implicit neural representations (INRs) use neural networks to provide continuous and resolution-independent representations of complex signals.<n>The proposed FKAN utilizes learnable activation functions modeled as Fourier series in the first layer to effectively control and learn the task-specific frequency components.<n> Experimental results show that our proposed FKAN model outperforms three state-of-the-art baseline schemes.
  arXiv  Detail & Related papers  (2024-09-14T05:53:33Z)
• Robust Fourier Neural Networks [1.0589208420411014]
  We show that introducing a simple diagonal layer after the Fourier embedding layer makes the network more robust to measurement noise. Under certain conditions, our proposed approach can also learn functions that are noisy mixtures of nonlinear functions of Fourier features.
  arXiv  Detail & Related papers  (2024-09-03T16:56:41Z)
• Decouple Graph Neural Networks: Train Multiple Simple GNNs Simultaneously Instead of One [60.5818387068983]
  Graph neural networks (GNN) suffer from severe inefficiency. We propose to decouple a multi-layer GNN as multiple simple modules for more efficient training. We show that the proposed framework is highly efficient with reasonable performance.
  arXiv  Detail & Related papers  (2023-04-20T07:21:32Z)
• Frame Flexible Network [52.623337134518835]
  Existing video recognition algorithms always conduct different training pipelines for inputs with different frame numbers. If we evaluate the model using other frames which are not used in training, we observe the performance will drop significantly. We propose a general framework, named Frame Flexible Network (FFN), which enables the model to be evaluated at different frames to adjust its computation.
  arXiv  Detail & Related papers  (2023-03-26T20:51:35Z)
• Incremental Spatial and Spectral Learning of Neural Operators for Solving Large-Scale PDEs [86.35471039808023]
  We introduce the Incremental Fourier Neural Operator (iFNO), which progressively increases the number of frequency modes used by the model. We show that iFNO reduces total training time while maintaining or improving generalization performance across various datasets. Our method demonstrates a 10% lower testing error, using 20% fewer frequency modes compared to the existing Fourier Neural Operator, while also achieving a 30% faster training.
  arXiv  Detail & Related papers  (2022-11-28T09:57:15Z)
• Transform Once: Efficient Operator Learning in Frequency Domain [69.74509540521397]
  We study deep neural networks designed to harness the structure in frequency domain for efficient learning of long-range correlations in space or time. This work introduces a blueprint for frequency domain learning through a single transform: transform once (T1)
  arXiv  Detail & Related papers  (2022-11-26T01:56:05Z)
• Functional Regularization for Reinforcement Learning via Learned Fourier Features [98.90474131452588]
  We propose a simple architecture for deep reinforcement learning by embedding inputs into a learned Fourier basis. We show that it improves the sample efficiency of both state-based and image-based RL.
  arXiv  Detail & Related papers  (2021-12-06T18:59:52Z)
• Factorized Fourier Neural Operators [77.47313102926017]
  The Factorized Fourier Neural Operator (F-FNO) is a learning-based method for simulating partial differential equations. We show that our model maintains an error rate of 2% while still running an order of magnitude faster than a numerical solver.
  arXiv  Detail & Related papers  (2021-11-27T03:34:13Z)
• Fourier Neural Networks as Function Approximators and Differential Equation Solvers [0.456877715768796]
  The choice of activation and loss function yields results that replicate a Fourier series expansion closely. We validate this FNN on naturally periodic smooth functions and on piecewise continuous periodic functions. The main advantages of the current approach are the validity of the solution outside the training region, interpretability of the trained model, and simplicity of use.
  arXiv  Detail & Related papers  (2020-05-27T00:30:58Z)

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.