The well-known work of Zhang et al. (2017) highlighted intriguing experimental phenomena about deep net training - specifically, optimization and generalization-and asked whether theory could explain them. They showed that sufficiently powerful nets (with vastly more parameters than number of training samples) can attain zero training error, regardless of whether the data is properly labeled or randomly labeled. Obviously, trainingwith randomly labeled data cannot generalize, whereas training with properly labeled data generalizes. See Figure 2 replicating some of these results. Recent papers have begun to provide explanations, showing that gradient descent can allow an overparametrized multi-layernet to attain arbitrarily low training error on fairly generic datasets (Du et al., 2018a,c; Li & Liang, 2018; Allen-Zhu et al., 2018b; Zou et al., 2018), provided the amount of overparametrization isa high polynomial of the relevant parameters (i.e.

Markov random fields (MRFs) find applications in a variety of machine learning areas, while the inference and learning of such models are challenging in general. In this paper, we propose the Adversarial Variational Inference and Learning (AVIL) algorithm to solve the problems with a minimal assumption about the model structure of an MRF. AVIL employs two variational distributions to approximately infer the latent variables and estimate the partition function, respectively. The variational distributions, which are parameterized as neural networks, provide an estimate of the negative log likelihood of the MRF. On one hand, the estimate is in an intuitive form of approximate contrastive free energy. On the other hand, the estimate is a minimax optimization problem, which is solved by stochastic gradient descent in an alternating manner. We apply AVIL to various undirected generative models in a fully black-box manner and obtain better results than existing competitors on several real datasets.

State-of-the-art neural networks are vulnerable to adversarial examples; they can easily misclassify inputs that are imperceptibly different than their training and test data. In this work, we establish that the use of cross-entropy loss function and the low-rank features of the training data have responsibility for the existence of these inputs. Based on this observation, we suggest that addressing adversarial examples requires rethinking the use of cross-entropy loss function and looking for an alternative that is more suited for minimization with low-rank features. In this direction, we present a training scheme called differential training, which uses a loss function defined on the differences between the features of points from opposite classes. We show that differential training can ensure a large margin between the decision boundary of the neural network and the points in the training dataset. This larger margin increases the amount of perturbation needed to flip the prediction of the classifier and makes it harder to find an adversarial example with small perturbations. We test differential training on a binary classification task with CIFAR-10 dataset and demonstrate that it radically reduces the ratio of images for which an adversarial example could be found -- not only in the training dataset, but in the test dataset as well.

We study the problem of minimizing a strongly convex and smooth function when we have noisy estimates of its gradient. We propose a novel multistage accelerated algorithm that is universally optimal in the sense that it achieves the optimal rate both in the deterministic and stochastic case and operates without knowledge of noise characteristics. The algorithm consists of stages that use a stochastic version of Nesterov's accelerated algorithm with a specific restart and parameters selected to achieve the fastest reduction in the bias-variance terms in the convergence rate bounds.

Rapid improvements in machine learning over the past decade are beginning to have far-reaching effects. For communications, engineers with limited domain expertise can now use off-the-shelf learning packages to design high-performance systems based on simulations. Prior to the current revolution in machine learning, the majority of communication engineers were quite aware that system parameters (such as filter coefficients) could be learned using stochastic gradient descent. It was not at all clear, however, that more complicated parts of the system architecture could be learned as well. In this paper, we discuss the application of machine-learning techniques to two communications problems and focus on what can be learned from the resulting systems. We were pleasantly surprised that the observed gains in one example have a simple explanation that only became clear in hindsight. In essence, deep learning discovered a simple and effective strategy that had not been considered earlier.

We present an online algorithm that induces decision trees using gradient information as the source of supervision. In contrast to previous approaches to gradient-based tree learning, we do not require soft splits or construction of a new tree for every update. In experiments, our method performs comparably to standard incremental classification trees and outperforms state of the art incremental regression trees. We also show how the method can be used to construct a novel type of neural network layer suited to learning representations from tabular data and find that it increases accuracy of multiclass and multi-label classification.

The weights of a neural network cannot be calculated using an analytical method. Instead, the weights must be discovered via an empirical optimization procedure called stochastic gradient descent.

Now, what was the Gradient Descent algorithm? Above algorithm says, to perform the GD, we need to calculate the gradient of the cost function J. And to calculate the gradient of the cost function, we need to sum (yellow circle!) the cost of each sample. If we have 3 million samples, we have to loop through 3 million times or use the dot product. Do you see np.dot(X.T, y_hat-y) above?

Recent studies on diffusion-based sampling methods have shown that Langevin Monte Carlo (LMC) algorithms can be beneficial for non-convex optimization, and rigorous theoretical guarantees have been proven for both asymptotic and finite-time regimes. Algorithmically, LMC-based algorithms resemble the well-known gradient descent (GD) algorithm, where the GD recursion is perturbed by an additive Gaussian noise whose variance has a particular form. Fractional Langevin Monte Carlo (FLMC) is a recently proposed extension of LMC, where the Gaussian noise is replaced by a heavy-tailed {\alpha}-stable noise. As opposed to its Gaussian counterpart, these heavy-tailed perturbations can incur large jumps and it has been empirically demonstrated that the choice of {\alpha}-stable noise can provide several advantages in modern machine learning problems, both in optimization and sampling contexts. However, as opposed to LMC, only asymptotic convergence properties of FLMC have been yet established. In this study, we analyze the non-asymptotic behavior of FLMC for non-convex optimization and prove finite-time bounds for its expected suboptimality. Our results show that the weak-error of FLMC increases faster than LMC, which suggests using smaller step-sizes in FLMC. We finally extend our results to the case where the exact gradients are replaced by stochastic gradients and show that similar results hold in this setting as well.

Radial-basis-function networks are traditionally defined for sets of vector-based observations. In this short paper, we reformulate such networks so that they can be applied to adjacency-matrix representations of weighted, directed graphs that represent the relationships between object pairs. We re-state the sum-of-squares objective function so that it is purely dependent on entries from the adjacency matrix. From this objective function, we derive a gradient descent update for the network weights. We also derive a gradient update that simulates the repositioning of the radial basis prototypes and changes in the radial basis prototype parameters. An important property of our radial basis function networks is that they are guaranteed to yield the same responses as conventional radial-basis networks trained on a corresponding vector realization of the relationships encoded by the adjacency-matrix. Such a vector realization only needs to provably exist for this property to hold, which occurs whenever the relationships correspond to distances from some arbitrary metric applied to a latent set of vectors. We therefore completely avoid needing to actually construct vectorial realizations via multi-dimensional scaling, which ensures that the underlying relationships are totally preserved.