ترغب بنشر مسار تعليمي؟ اضغط هنا

Training (Overparametrized) Neural Networks in Near-Linear Time

97   0   0.0 ( 0 )
 نشر من قبل Binghui Peng
 تاريخ النشر 2020
  مجال البحث الهندسة المعلوماتية
والبحث باللغة English




اسأل ChatGPT حول البحث

The slow convergence rate and pathological curvature issues of first-order gradient methods for training deep neural networks, initiated an ongoing effort for developing faster $mathit{second}$-$mathit{order}$ optimization algorithms beyond SGD, without compromising the generalization error. Despite their remarkable convergence rate ($mathit{independent}$ of the training batch size $n$), second-order algorithms incur a daunting slowdown in the $mathit{cost}$ $mathit{per}$ $mathit{iteration}$ (inverting the Hessian matrix of the loss function), which renders them impractical. Very recently, this computational overhead was mitigated by the works of [ZMG19,CGH+19}, yielding an $O(mn^2)$-time second-order algorithm for training two-layer overparametrized neural networks of polynomial width $m$. We show how to speed up the algorithm of [CGH+19], achieving an $tilde{O}(mn)$-time backpropagation algorithm for training (mildly overparametrized) ReLU networks, which is near-linear in the dimension ($mn$) of the full gradient (Jacobian) matrix. The centerpiece of our algorithm is to reformulate the Gauss-Newton iteration as an $ell_2$-regression problem, and then use a Fast-JL type dimension reduction to $mathit{precondition}$ the underlying Gram matrix in time independent of $M$, allowing to find a sufficiently good approximate solution via $mathit{first}$-$mathit{order}$ conjugate gradient. Our result provides a proof-of-concept that advanced machinery from randomized linear algebra -- which led to recent breakthroughs in $mathit{convex}$ $mathit{optimization}$ (ERM, LPs, Regression) -- can be carried over to the realm of deep learning as well.



قيم البحث

اقرأ أيضاً

We introduce a pruning algorithm that provably sparsifies the parameters of a trained model in a way that approximately preserves the models predictive accuracy. Our algorithm uses a small batch of input points to construct a data-informed importance sampling distribution over the networks parameters, and adaptively mixes a sampling-based and deterministic pruning procedure to discard redundant weights. Our pruning method is simultaneously computationally efficient, provably accurate, and broadly applicable to various network architectures and data distributions. Our empirical comparisons show that our algorithm reliably generates highly compressed networks that incur minimal loss in performance relative to that of the original network. We present experimental results that demonstrate our algorithms potential to unearth essential network connections that can be trained successfully in isolation, which may be of independent interest.
It is well-known that overparametrized neural networks trained using gradient-based methods quickly achieve small training error with appropriate hyperparameter settings. Recent papers have proved this statement theoretically for highly overparametri zed networks under reasonable assumptions. These results either assume that the activation function is ReLU or they crucially depend on the minimum eigenvalue of a certain Gram matrix depending on the data, random initialization and the activation function. In the later case, existing works only prove that this minimum eigenvalue is non-zero and do not provide quantitative bounds. On the empirical side, a contemporary line of investigations has proposed a number of alternative activation functions which tend to perform better than ReLU at least in some settings but no clear understanding has emerged. This state of affairs underscores the importance of theoretically understanding the impact of activation functions on training. In the present paper, we provide theoretical results about the effect of activation function on the training of highly overparametrized 2-layer neural networks. A crucial property that governs the performance of an activation is whether or not it is smooth. For non-smooth activations such as ReLU, SELU and ELU, all eigenvalues of the associated Gram matrix are large under minimal assumptions on the data. For smooth activations such as tanh, swish and polynomials, the situation is more complex. If the subspace spanned by the data has small dimension then the minimum eigenvalue of the Gram matrix can be small leading to slow training. But if the dimension is large and the data satisfies another mild condition, then the eigenvalues are large. If we allow deep networks, then the small data dimension is not a limitation provided that the depth is sufficient. We discuss a number of extensions and applications of these results.
Convolutional neural networks have achieved astonishing results in different application areas. Various methods that allow us to use these models on mobile and embedded devices have been proposed. Especially binary neural networks are a promising app roach for devices with low computational power. However, training accurate binary models from scratch remains a challenge. Previous work often uses prior knowledge from full-precision models and complex training strategies. In our work, we focus on increasing the performance of binary neural networks without such prior knowledge and a much simpler training strategy. In our experiments we show that we are able to achieve state-of-the-art results on standard benchmark datasets. Further, to the best of our knowledge, we are the first to successfully adopt a network architecture with dense connections for binary networks, which lets us improve the state-of-the-art even further.
Differentially private stochastic gradient descent (DPSGD) is a variation of stochastic gradient descent based on the Differential Privacy (DP) paradigm which can mitigate privacy threats arising from the presence of sensitive information in training data. One major drawback of training deep neural networks with DPSGD is a reduction in the models accuracy. In this paper, we propose an alternative method for preserving data privacy based on introducing noise through learnable probability distributions, which leads to a significant improvement in the utility of the resulting private models. We also demonstrate that normalization layers have a large beneficial impact on the performance of deep neural networks with noisy parameters. In particular, we show that contrary to general belief, a large amount of random noise can be added to the weights of neural networks without harming the performance, once the networks are augmented with normalization layers. We hypothesize that this robustness is a consequence of the scale invariance property of normalization operators. Building on these observations, we propose a new algorithmic technique for training deep neural networks under very low privacy budgets by sampling weights from Gaussian distributions and utilizing batch or layer normalization techniques to prevent performance degradation. Our method outperforms previous approaches, including DPSGD, by a substantial margin on a comprehensive set of experiments on Computer Vision and Natural Language Processing tasks. In particular, we obtain a 20 percent accuracy improvement over DPSGD on the MNIST and CIFAR10 datasets with DP-privacy budgets of $varepsilon = 0.05$ and $varepsilon = 2.0$, respectively. Our code is available online: https://github.com/uds-lsv/SIDP.
Todays deep learning models are primarily trained on CPUs and GPUs. Although these models tend to have low error, they consume high power and utilize large amount of memory owing to double precision floating point learning parameters. Beyond the Moor es law, a significant portion of deep learning tasks would run on edge computing systems, which will form an indispensable part of the entire computation fabric. Subsequently, training deep learning models for such systems will have to be tailored and adopted to generate models that have the following desirable characteristics: low error, low memory, and low power. We believe that deep neural networks (DNNs), where learning parameters are constrained to have a set of finite discrete values, running on neuromorphic computing systems would be instrumental for intelligent edge computing systems having these desirable characteristics. To this extent, we propose the Combinatorial Neural Network Training Algorithm (CoNNTrA), that leverages a coordinate gradient descent-based approach for training deep learning models with finite discrete learning parameters. Next, we elaborate on the theoretical underpinnings and evaluate the computational complexity of CoNNTrA. As a proof of concept, we use CoNNTrA to train deep learning models with ternary learning parameters on the MNIST, Iris and ImageNet data sets and compare their performance to the same models trained using Backpropagation. We use following performance metrics for the comparison: (i) Training error; (ii) Validation error; (iii) Memory usage; and (iv) Training time. Our results indicate that CoNNTrA models use 32x less memory and have errors at par with the Backpropagation models.

الأسئلة المقترحة

التعليقات
جاري جلب التعليقات جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
mircosoft-partner

هل ترغب بارسال اشعارات عن اخر التحديثات في شمرا-اكاديميا