CSNNs: Unsupervised, Backpropagation-free Convolutional Neural Networks for Representation Learning
University of Applied Sciences Kempten（肯普滕应用科学大学）
Bonifaz Stuhr, Jürgen Brauer
备注：
18 pages,18 figures, Author's extended version of the paper. Final version presented at 18th IEEE International Conference on Machine Learning and Applications (ICMLA). Boca Raton, Florida / USA. 2019
This work combines Convolutional Neural Networks (CNNs), clustering via Self-Organizing Maps (SOMs) and Hebbian Learning to propose the building blocks of Convolutional Self-Organizing Neural Networks (CSNNs), which learn representations in an unsupervised and Backpropagation-free manner. Our approach replaces the learning of traditional convolutional layers from CNNs with the competitive learning procedure of SOMs and simultaneously learns local masks between those layers with separate Hebbian-like learning rules to overcome the problem of disentangling factors of variation when filters are learned through clustering. We investigate the learned representation by designing two simple models with our building blocks, achieving comparable performance to many methods which use Backpropagation, while we reach comparable performance on Cifar10 and give baseline performances on Cifar100, Tiny ImageNet and a small subset of ImageNet for Backpropagation-free methods.
Spiking Two-Stream Methods with Unsupervised STDP-based Learning for Action Recognition
Univ. Lille（里尔大学）；CNRS（法国国家科学研究中心）；Centrale Lille（里尔中央理工学院）；MR 9189 -- CRIStAL -- Centre de Recherche en Informatique, Signal et Automatique de Lille（里尔信息、信号与自动化研究中心（MR 9189 - CRIStAL））
Amirkabir University of Technology（阿米尔卡比尔理工大学）；Shahid Beheshti University（沙希德·贝赫什提大学）；CNRS Université Toulouse 3（法国国家科学研究中心图卢兹第三大学）
We recently proposed the STiDi-BP algorithm, which avoids backward recursive gradient computation, for training multi-layer spiking neural networks (SNNs) with single-spike-based temporal coding. The algorithm employs a linear approximation to compute the derivative of the spike latency with respect to the membrane potential and it uses spiking neurons with piecewise linear postsynaptic potential to reduce the computational cost and the complexity of neural processing. In this paper, we extend the STiDi-BP algorithm to employ it in deeper and convolutional architectures. The evaluation results…
Training deep neural networks with the error backpropagation algorithm is considered implausible from a biological perspective. Numerous recent publications suggest elaborate models for biologically plausible variants of deep learning, typically defining success as reaching around 98% test accuracy on the MNIST data set. Here, we investigate how far we can go on digit (MNIST) and object (CIFAR10) classification with biologically plausible, local learning rules in a network with one hidden layer and a single readout layer. The hidden layer weights are either fixed (random or random Gabor filter…
Deep neural nets typically perform end-to-end backpropagation to learn the weights, a procedure that creates synchronization constraints in the weight update step across layers and is not biologically plausible. Recent advances in unsupervised contrastive representation learning point to the question of whether a learning algorithm can also be made local, that is, the updates of lower layers do not directly depend on the computation of upper layers. While Greedy InfoMax separately learns each block with a local objective, we found that it consistently hurts readout accuracy in state-of-the-art…
University of Geneva（日内瓦大学）；University of Applied Sciences（应用科学大学）；Apple Inc.（苹果公司）
Modern neural network training relies on piece-wise (sub-)differentiable functions in order to use backpropagation to update model parameters. In this work, we introduce a novel method to allow simple non-differentiable functions at intermediary layers of deep neural networks. We do so by training with a differentiable approximation bridge (DAB) neural network which approximates the non-differentiable forward function and provides gradient updates during backpropagation. We present strong empirical results (performing over 600 experiments) in four different domains: unsupervised (image) repres…
Unsupervised End-to-End Training with a Self-Defined Target
Laboratoire Albert Fert（阿尔贝·费尔实验室）；CNRS（法国国家科学研究中心）；Thales（泰雷兹集团）；Université Paris-Saclay（巴黎萨克雷大学）；Centre de Nanosciences et de Nanotechnologies（纳米科学与纳米技术中心）
University of Chicago（芝加哥大学）；Department of Statistics（统计系）
We develop biologically plausible training mechanisms for self-supervised learning (SSL) in deep networks. Specifically, by biological plausible training we mean (i) All updates of weights are based on current activities of pre-synaptic units and current, or activity retrieved from short term memory of post synaptic units, including at the top-most error computing layer, (ii) Complex computations such as normalization, inner products and division are avoided (iii) Asymmetric connections between units, (iv) Most learning is carried out in an unsupervised manner. SSL with a contrastive loss sati…
MIT-IBM Watson AI Lab（麻省理工-IBM沃森人工智能实验室）；IBM Research（IBM研究院）；Institute for Advanced Study（高等研究院）；Princeton Neuroscience Institute（普林斯顿神经科学研究所）；Princeton University（普林斯顿大学）
It is widely believed that the backpropagation algorithm is essential for learning good feature detectors in early layers of artificial neural networks, so that these detectors are useful for the task performed by the higher layers of that neural network. At the same time, the traditional form of backpropagation is biologically implausible. In the present paper we propose an unusual learning rule, which has a degree of biological plausibility, and which is motivated by Hebb's idea that change of the synapse strength should be local - i.e. should depend only on the activities of the pre and pos…
Data61, CSIRO（澳大利亚联邦科学与工业研究组织Data61）；Australian National University（澳大利亚国立大学）
This paper introduces an extension of the backpropagation algorithm that enables us to have layers with constrained weights in a deep network. In particular, we make use of the Riemannian geometry and optimization techniques on matrix manifolds to step outside of normal practice in training deep networks, equipping the network with structures such as orthogonality or positive definiteness. Based on our development, we make another contribution by introducing the Stiefel layer, a layer with orthogonal weights. Among various applications, Stiefel layers can be used to design orthogonal filter ba…
Biological image processing is performed by complex neural networks composed of thousands of neurons interconnected via thousands of synapses, some of which are excitatory and others inhibitory. Spiking neural models are distinguished from classical neurons by being biological plausible and exhibiting the same dynamics as those observed in biological neurons. This paper proposes a Natural Convolutional Neural Network (NatCSNN) which is a 3-layer bio-inspired Convolutional Spiking Neural Network (CSNN), for classifying objects extracted from natural images. A two-stage training algorithm is pro…
Despite its great success, backpropagation has certain limitations that necessitate the investigation of new learning methods. In this study, we present a biologically plausible local learning rule that improves upon Hebb's well-known proposal and discovers unsupervised features by local competitions among neurons. This simple learning rule enables the creation of a forward learning paradigm called activation learning, in which the output activation (sum of the squared output) of the neural network estimates the likelihood of the input patterns, or "learn more, activate more" in simpler terms.…
Rochester Institute of Technology（罗切斯特理工学院）；AI Research Lab（AI研究实验室）；University of Washington（华盛顿大学）；Rochester, New York, USA（纽约罗切斯特市）；Los Angeles, California, USA（加州洛杉矶市）；Seattle, Washington, USA（华盛顿州西雅图市）
Center for Advanced Intelligence Project (AIP), RIKEN（理化学研究所先进智能项目中心）；Graduate School of Informatics, Kyoto University（京都大学信息学研究科）；National Institute of Advanced Industrial Science and Technology (AIST)（日本产业技术综合研究所）；LTCI, Télécom Paris, Institut Polytechnique de Paris（巴黎理工学院巴黎电信学院信息处理与通信实验室）
This paper describes the practical response- and performance-aware development of online speech enhancement for an augmented reality (AR) headset that helps a user understand conversations made in real noisy echoic environments (e.g., cocktail party). One may use a state-of-the-art blind source separation method called fast multichannel nonnegative matrix factorization (FastMNMF) that works well in various environments thanks to its unsupervised nature. Its heavy computational cost, however, prevents its application to real-time processing. In contrast, a supervised beamforming method that use…
Traditional methods for motion estimation estimate the motion field F between a pair of images as the one that minimizes a predesigned cost function. In this paper, we propose a direct method and train a Convolutional Neural Network (CNN) that when, at test time, is given a pair of images as input it produces a dense motion field F at its output layer. In the absence of large datasets with ground truth motion that would allow classical supervised training, we propose to train the network in an unsupervised manner. The proposed cost function that is optimized during training, is based on the cl…
Interpretable and backpropagation-free Green Learning for efficient multi-task echocardiographic segmentation and classification
Harvard Medical School and Massachusetts General Hospital（哈佛医学院和麻省总医院）；Graduate Institute of Biomedical Electronics and Bioinformatics（生物医学电子与生物信息学研究生院）；University of Southern California（南加州大学）
This paper proposes a method for machine learning from unlabeled data in the form of a time-series. The mapping that is learned is shown to extract slowly evolving information that would be useful for control applications, while efficiently filtering out unwanted, higher-frequency noise. The method consists of training a feedforward artificial neural network with backpropagation using two opposing objectives. The first of these is to minimize the squared changes in activations between time steps of each unit in the network. This "temporal smoothing" has the effect of correlating inputs that oc…
University of Florida（佛罗里达大学）；Universidad de los Andes（安第斯大学）
We propose a convolutional recurrent neural network, with Winner-Take-All dropout for high dimensional unsupervised feature learning in multi-dimensional time series. We apply the proposedmethod for object recognition with temporal context in videos and obtain better results than comparable methods in the literature, including the Deep Predictive Coding Networks previously proposed by Chalasani and Principe.Our contributions can be summarized as a scalable reinterpretation of the Deep Predictive Coding Networks trained end-to-end with backpropagation through time, an extension of the previousl…
Victoria University of Wellington（惠灵顿维多利亚大学）；ETH Zürich（苏黎世联邦理工学院）
In this paper, we propose a novel unsupervised domain adaptation algorithm based on deep learning for visual object recognition. Specifically, we design a new model called Deep Reconstruction-Classification Network (DRCN), which jointly learns a shared encoding representation for two tasks: i) supervised classification of labeled source data, and ii) unsupervised reconstruction of unlabeled target data.In this way, the learnt representation not only preserves discriminability, but also encodes useful information from the target domain. Our new DRCN model can be optimized by using backpropagati…
Psychological studies have found that human visual tracking system involves learning, memory, and planning. Despite recent successes, not many works have focused on memory and planning in deep learning based tracking. We are thus interested in memory augmented network, where an external memory remembers the evolving appearance of the target (foreground) object without backpropagation for updating weights. Our Dual Augmented Memory Network (DAWN) is unique in remembering both target and background, and using an improved attention LSTM memory to guide the focus on memorized features. DAWN is eff…
Meinig School of Biomedical Engineering, Cornell University（康奈尔大学迈尼奇生物医学工程学院）；Weill Cornell Medicine（威尔康奈尔医学院）；Northern Jiangsu People's Hospital（苏北人民医院）
Purpose: To use a deep neural network (DNN) for solving the optimization problem of water/fat separation and to compare supervised and unsupervised training. Methods: The current T2*-IDEAL algorithm for solving fat/water separation is dependent on initialization. Recently, deep neural networks (DNN) have been proposed to solve fat/water separation without the need for suitable initialization. However, this approach requires supervised training of DNN (STD) using the reference fat/water separation images. Here we propose two novel DNN water/fat separation methods 1) unsupervised training of DNN…
Recently, nested dropout was proposed as a method for ordering representation units in autoencoders by their information content, without diminishing reconstruction cost. However, it has only been applied to training fully-connected autoencoders in an unsupervised setting. We explore the impact of nested dropout on the convolutional layers in a CNN trained by backpropagation, investigating whether nested dropout can provide a simple and systematic way to determine the optimal representation size with respect to the desired accuracy and desired task and data complexity.
Imperial College London（伦敦帝国理工学院）；ETH Zurich（苏黎世联邦理工学院）；University of Cambridge（剑桥大学）
We introduce gvnn, a neural network library in Torch aimed towards bridging the gap between classic geometric computer vision and deep learning. Inspired by the recent success of Spatial Transformer Networks, we propose several new layers which are often used as parametric transformations on the data in geometric computer vision. These layers can be inserted within a neural network much in the spirit of the original spatial transformers and allow backpropagation to enable end-to-end learning of a network involving any domain knowledge in geometric computer vision. This opens up applications in…
IBM Research（IBM研究院）；Princeton Neuroscience Institute（普林斯顿神经科学研究所）；Princeton University（普林斯顿大学）；MIT-IBM Watson AI Lab（MIT-IBM沃森人工智能实验室）
Local Hebbian learning is believed to be inferior in performance to end-to-end training using a backpropagation algorithm. We question this popular belief by designing a local algorithm that can learn convolutional filters at scale on large image datasets. These filters combined with patch normalization and very steep non-linearities result in a good classification accuracy for shallow networks trained locally, as opposed to end-to-end. The filters learned by our algorithm contain both orientation selective units and unoriented color units, resembling the responses of pyramidal neurons located…
Convolutional neural networks (CNNs) have had great success in many real-world applications and have also been used to model visual processing in the brain. However, these networks are quite brittle - small changes in the input image can dramatically change a network's output prediction. In contrast to what is known from biology, these networks largely rely on feedforward connections, ignoring the influence of recurrent connections. They also focus on supervised rather than unsupervised learning. To address these issues, we combine traditional supervised learning via backpropagation with a spe…
Inria（法国国家信息与自动化研究所）；Laboratoire Jean Kuntzmann（让·昆茨曼实验室）；CNRS（法国国家科学研究中心）；Univ. Grenoble Alpes（格勒诺布尔阿尔卑斯大学）
In this paper, we introduce a new image representation based on a multilayer kernel machine. Unlike traditional kernel methods where data representation is decoupled from the prediction task, we learn how to shape the kernel with supervision. We proceed by first proposing improvements of the recently-introduced convolutional kernel networks (CKNs) in the context of unsupervised learning; then, we derive backpropagation rules to take advantage of labeled training data. The resulting model is a new type of convolutional neural network, where optimizing the filters at each layer is equivalent to …
Learnable keypoint detectors and descriptors are beginning to outperform classical hand-crafted feature extraction methods. Recent studies on self-supervised learning of visual representations have driven the increasing performance of learnable models based on deep networks. By leveraging traditional data augmentations and homography transformations, these networks learn to detect corners under adverse conditions such as extreme illumination changes. However, their generalization capabilities are limited to corner-like features detected a priori by classical methods or synthetically generated …
Communication University of China（中国传媒大学）；Neuroscience and Intelligent Media Institute（神经科学与智能媒体研究院）
Backpropagation-based supervised learning has achieved great success in computer vision tasks. However, its biological plausibility is always controversial. Recently, the bio-inspired Hebbian learning rule (HLR) has received extensive attention. Self-Organizing Map (SOM) uses the competitive HLR to establish connections between neurons, obtaining visual features in an unsupervised way. Although the representation of SOM neurons shows some brain-like characteristics, it is still quite different from the neuron representation in the human visual cortex. This paper proposes an improved SOM with m…
Kavli Institute for Systems Neuroscience（卡夫利系统神经科学研究所）；NTNU（挪威科技大学）；Max-Planck-Insitute for Human Cognitive and Brain Sciences（马克斯·普朗克人类认知与脑科学研究所）；UCL（伦敦大学学院）
Deep artificial neural networks (DNNs) trained through backpropagation provide effective models of the mammalian visual system, accurately capturing the hierarchy of neural responses through primary visual cortex to inferior temporal cortex (IT). However, the ability of these networks to explain representations in higher cortical areas is relatively lacking and considerably less well researched. For example, DNNs have been less successful as a model of the egocentric to allocentric transformation embodied by circuits in retrosplenial and posterior parietal cortex. We describe a novel scene per…
