Towards training digitally-tied analog blocks via hybrid gradient computation

• URL: http://arxiv.org/abs/2409.03306v1
• Date: Thu, 5 Sep 2024 07:22:19 GMT
• Title: Towards training digitally-tied analog blocks via hybrid gradient computation
• Authors: Timothy Nest, Maxence Ernoult,
• Abstract summary: We introduce Feedforward-tied Energy-based Models (ff-EBMs) We derive a novel algorithm to compute gradients end-to-end in ff-EBMs by backpropagating and "eq-propagating" through feedforward and energy-based parts respectively. Our approach offers a principled, scalable, and incremental roadmap to gradually integrate self-trainable analog computational primitives into existing digital accelerators.
• Score: 1.800676987432211
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Power efficiency is plateauing in the standard digital electronics realm such that novel hardware, models, and algorithms are needed to reduce the costs of AI training. The combination of energy-based analog circuits and the Equilibrium Propagation (EP) algorithm constitutes one compelling alternative compute paradigm for gradient-based optimization of neural nets. Existing analog hardware accelerators, however, typically incorporate digital circuitry to sustain auxiliary non-weight-stationary operations, mitigate analog device imperfections, and leverage existing digital accelerators.This heterogeneous hardware approach calls for a new theoretical model building block. In this work, we introduce Feedforward-tied Energy-based Models (ff-EBMs), a hybrid model comprising feedforward and energy-based blocks accounting for digital and analog circuits. We derive a novel algorithm to compute gradients end-to-end in ff-EBMs by backpropagating and "eq-propagating" through feedforward and energy-based parts respectively, enabling EP to be applied to much more flexible and realistic architectures. We experimentally demonstrate the effectiveness of the proposed approach on ff-EBMs where Deep Hopfield Networks (DHNs) are used as energy-based blocks. We first show that a standard DHN can be arbitrarily split into any uniform size while maintaining performance. We then train ff-EBMs on ImageNet32 where we establish new SOTA performance in the EP literature (46 top-1 %). Our approach offers a principled, scalable, and incremental roadmap to gradually integrate self-trainable analog computational primitives into existing digital accelerators.

Related papers

• A Fully Hardware Implemented Accelerator Design in ReRAM Analog Computing without ADCs [5.6496088684920345]
  ReRAM-based accelerators process neural networks via analog Computing-in-Memory (CiM) for ultra-high energy efficiency. This work explores the hardware implementation of the Sigmoid and SoftMax activation functions of neural networks with crossbarally binarized neurons. We propose a complete ReRAM-based Analog Computing Accelerator (RACA) that accelerates neural network computation by leveraging inferenceally binarized neurons.
  arXiv  Detail & Related papers  (2024-12-27T09:38:19Z)
• Synergistic Development of Perovskite Memristors and Algorithms for Robust Analog Computing [53.77822620185878]
  We propose a synergistic methodology to concurrently optimize perovskite memristor fabrication and develop robust analog DNNs. We develop "BayesMulti", a training strategy utilizing BO-guided noise injection to improve the resistance of analog DNNs to memristor imperfections. Our integrated approach enables use of analog computing in much deeper and wider networks, achieving up to 100-fold improvements.
  arXiv  Detail & Related papers  (2024-12-03T19:20:08Z)
• EKGNet: A 10.96{\mu}W Fully Analog Neural Network for Intra-Patient Arrhythmia Classification [79.7946379395238]
  We present an integrated approach by combining analog computing and deep learning for electrocardiogram (ECG) arrhythmia classification. We propose EKGNet, a hardware-efficient and fully analog arrhythmia classification architecture that archives high accuracy with low power consumption.
  arXiv  Detail & Related papers  (2023-10-24T02:37:49Z)
• Energy Efficient Hardware Acceleration of Neural Networks with Power-of-Two Quantisation [0.0]
  We show that a hardware neural network accelerator with PoT weights implemented on the Zynq UltraScale + MPSoC ZCU104 FPGA can be at least $1.4x$ more energy efficient than the uniform quantisation version.
  arXiv  Detail & Related papers  (2022-09-30T06:33:40Z)
• Decomposition of Matrix Product States into Shallow Quantum Circuits [62.5210028594015]
  tensor network (TN) algorithms can be mapped to parametrized quantum circuits (PQCs) We propose a new protocol for approximating TN states using realistic quantum circuits. Our results reveal one particular protocol, involving sequential growth and optimization of the quantum circuit, to outperform all other methods.
  arXiv  Detail & Related papers  (2022-09-01T17:08:41Z)
• Pretraining Graph Neural Networks for few-shot Analog Circuit Modeling and Design [68.1682448368636]
  We present a supervised pretraining approach to learn circuit representations that can be adapted to new unseen topologies or unseen prediction tasks. To cope with the variable topological structure of different circuits we describe each circuit as a graph and use graph neural networks (GNNs) to learn node embeddings. We show that pretraining GNNs on prediction of output node voltages can encourage learning representations that can be adapted to new unseen topologies or prediction of new circuit level properties.
  arXiv  Detail & Related papers  (2022-03-29T21:18:47Z)
• Neural Network Training with Asymmetric Crosspoint Elements [1.0773924713784704]
  asymmetric conductance modulation of practical resistive devices critically degrades the classification of networks trained with conventional algorithms. Here, we describe and experimentally demonstrate an alternative fully-parallel training algorithm: Hamiltonian Descent. We provide critical intuition on why device asymmetry is fundamentally incompatible with conventional training algorithms and how the new approach exploits it as a useful feature instead.
  arXiv  Detail & Related papers  (2022-01-31T17:41:36Z)
• Prospects for Analog Circuits in Deep Networks [14.280112591737199]
  Operations typically used in machine learning al-gorithms can be implemented by compact analog circuits. With the recent advances in deep learning algorithms, focus has shifted to hardware digital accelerator designs. This paper presents abrief review of analog designs that implement various machine learning algorithms.
  arXiv  Detail & Related papers  (2021-06-23T14:49:21Z)
• Adaptive pruning-based optimization of parameterized quantum circuits [62.997667081978825]
  Variisy hybrid quantum-classical algorithms are powerful tools to maximize the use of Noisy Intermediate Scale Quantum devices. We propose a strategy for such ansatze used in variational quantum algorithms, which we call "Efficient Circuit Training" (PECT) Instead of optimizing all of the ansatz parameters at once, PECT launches a sequence of variational algorithms.
  arXiv  Detail & Related papers  (2020-10-01T18:14:11Z)
• Training End-to-End Analog Neural Networks with Equilibrium Propagation [64.0476282000118]
  We introduce a principled method to train end-to-end analog neural networks by gradient descent. We show mathematically that a class of analog neural networks (called nonlinear resistive networks) are energy-based models. Our work can guide the development of a new generation of ultra-fast, compact and low-power neural networks supporting on-chip learning.
  arXiv  Detail & Related papers  (2020-06-02T23:38:35Z)

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.