اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدOn the Convergence of Continuous Constrained Optimization for Structure Learning
185
0
0.0
(
0
)
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
Structure learning of directed acyclic graphs (DAGs) is a fundamental problem in many scientific endeavors. A new line of work, based on NOTEARS (Zheng et al., 2018), reformulates the structure learning problem as a continuous optimization one by leveraging an algebraic characterization of DAG constraint. The constrained problem is typically solved using the augmented Lagrangian method (ALM) which is often preferred to the quadratic penalty method (QPM) by virtue of its convergence result that does not require the penalty coefficient to go to infinity, hence avoiding ill-conditioning. In this work, we review the standard convergence result of the ALM and show that the required conditions are not satisfied in the recent continuous constrained formulation for learning DAGs. We demonstrate empirically that its behavior is akin to that of the QPM which is prone to ill-conditioning, thus motivating the use of second-order method in this setting. We also establish the convergence guarantee of QPM to a DAG solution, under mild conditions, based on a property of the DAG constraint term.
قيم البحث
اقرأ أيضاً
In deep learning with differential privacy (DP), the neural network achieves the privacy usually at the cost of slower convergence (and thus lower performance) than its non-private counterpart. This work gives the first convergence analysis of the DP
deep learning, through the lens of training dynamics and the neural tangent kernel (NTK). Our convergence theory successfully characterizes the effects of two key components in the DP training: the per-sample clipping (flat or layerwise) and the noise addition. Our analysis not only initiates a general principled framework to understand the DP deep learning with any network architecture and loss function, but also motivates a new clipping method -- the global clipping, that significantly improves the convergence while preserving the same privacy guarantee as the existing local clipping. In terms of theoretical results, we establish the precise connection between the per-sample clipping and NTK matrix. We show that in the gradient flow, i.e., with infinitesimal learning rate, the noise level of DP optimizers does not affect the convergence. We prove that DP gradient descent (GD) with global clipping guarantees the monotone convergence to zero loss, which can be violated by the existing DP-GD with local clipping. Notably, our analysis framework easily extends to other optimizers, e.g., DP-Adam. Empirically speaking, DP optimizers equipped with global clipping perform strongly on a wide range of classification and regression tasks. In particular, our global clipping is surprisingly effective at learning calibrated classifiers, in contrast to the existing DP classifiers which are oftentimes over-confident and unreliable. Implementation-wise, the new clipping can be realized by adding one line of code into the Opacus library.
This paper presents a new conditional GAN (named convex relaxing CGAN or crCGAN) to replicate the conventional constrained topology optimization algorithms in an extremely effective and efficient process. The proposed crCGAN consists of a generator a
nd a discriminator, both of which are deep convolutional neural networks (CNN) and the topology design constraint can be conditionally set to both the generator and discriminator. In order to improve the training efficiency and accuracy due to the dependency between the training images and the condition, a variety of crCGAN formulation are introduced to relax the non-convex design space. These new formulations were evaluated and validated via a series of comprehensive experiments. Moreover, a minibatch discrimination technique was introduced in the crCGAN training process to stabilize the convergence and avoid the mode collapse problems. Additional verifications were conducted using the state-of-the-art MNIST digits and CIFAR-10 images conditioned by class labels. The experimental evaluations clearly reveal that the new objective formulation with the minibatch discrimination training provides not only the accuracy but also the consistency of the designs.
When training end-to-end learned models for lossy compression, one has to balance the rate and distortion losses. This is typically done by manually setting a tradeoff parameter $beta$, an approach called $beta$-VAE. Using this approach it is difficu
lt to target a specific rate or distortion value, because the result can be very sensitive to $beta$, and the appropriate value for $beta$ depends on the model and problem setup. As a result, model comparison requires extensive per-model $beta$-tuning, and producing a whole rate-distortion curve (by varying $beta$) for each model to be compared. We argue that the constrained optimization method of Rezende and Viola, 2018 is a lot more appropriate for training lossy compression models because it allows us to obtain the best possible rate subject to a distortion constraint. This enables pointwise model comparisons, by training two models with the same distortion target and comparing their rate. We show that the method does manage to satisfy the constraint on a realistic image compression task, outperforms a constrained optimization method based on a hinge-loss, and is more practical to use for model selection than a $beta$-VAE.
The communication between data-generating devices is partially responsible for a growing portion of the worlds power consumption. Thus reducing communication is vital, both, from an economical and an ecological perspective. For machine learning, on-d
evice learning avoids sending raw data, which can reduce communication substantially. Furthermore, not centralizing the data protects privacy-sensitive data. However, most learning algorithms require hardware with high computation power and thus high energy consumption. In contrast, ultra-low-power processors, like FPGAs or micro-controllers, allow for energy-efficient learning of local models. Combined with communication-efficient distributed learning strategies, this reduces the overall energy consumption and enables applications that were yet impossible due to limited energy on local devices. The major challenge is then, that the low-power processors typically only have integer processing capabilities. This paper investigates an approach to communication-efficient on-device learning of integer exponential families that can be executed on low-power processors, is privacy-preserving, and effectively minimizes communication. The empirical evaluation shows that the approach can reach a model quality comparable to a centrally learned regular model with an order of magnitude less communication. Comparing the overall energy consumption, this reduces the required energy for solving the machine learning task by a significant amount.
The extrapolation strategy raised by Nesterov, which can accelerate the convergence rate of gradient descent methods by orders of magnitude when dealing with smooth convex objective, has led to tremendous success in training machine learning tasks. I
n this article, the convergence of individual iterates of projected subgradient (PSG) methods for nonsmooth convex optimization problems is theoretically studied based on Nesterovs extrapolation, which we name individual convergence. We prove that Nesterovs extrapolation has the strength to make the individual convergence of PSG optimal for nonsmooth problems. In light of this consideration, a direct modification of the subgradient evaluation suffices to achieve optimal individual convergence for strongly convex problems, which can be regarded as making an interesting step toward the open question about stochastic gradient descent (SGD) posed by Shamir. Furthermore, we give an extension of the derived algorithms to solve regularized learning tasks with nonsmooth losses in stochastic settings. Compared with other state-of-the-art nonsmooth methods, the derived algorithms can serve as an alternative to the basic SGD especially in coping with machine learning problems, where an individual output is needed to guarantee the regularization structure while keeping an optimal rate of convergence. Typically, our method is applicable as an efficient tool for solving large-scale $l$1-regularized hinge-loss learning problems. Several comparison experiments demonstrate that our individual output not only achieves an optimal convergence rate but also guarantees better sparsity than the averaged solution.
الأسئلة المقترحة
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
