A key appeal of the recently proposed Neural Ordinary Differential Equation (ODE) framework is that it seems to provide a continuous-time extension of discrete residual neural networks. As we show herein, though, trained Neural ODE models actually depend on the specific numerical method used during training. If the trained model is supposed to be a flow generated from an ODE, it should be possible to choose another numerical solver with equal or smaller numerical error without loss of performance. We observe that if training relies on a solver with overly coarse discretization, then testing with another solver of equal or smaller numerical error results in a sharp drop in accuracy. In such cases, the combination of vector field and numerical method cannot be interpreted as a flow generated from an ODE, which arguably poses a fatal breakdown of the Neural ODE concept. We observe, however, that there exists a critical step size beyond which the training yields a valid ODE vector field. We propose a method that monitors the behavior of the ODE solver during training to adapt its step size, aiming to ensure a valid ODE without unnecessarily increasing computational cost. We verify this adaption algorithm on two common bench mark datasets as well as a synthetic dataset. Furthermore, we introduce a novel synthetic dataset in which the underlying ODE directly generates a classification task.

Although recent works have brought some insights into the performance improvement of techniques used in state-of-the-art deep-learning models, more work is needed to understand their generalization properties. We shed light on this matter by linking the loss function to the output's sensitivity to its input. We find a rather strong empirical relation between the output sensitivity and the variance in the bias-variance decomposition of the loss function, which hints on using sensitivity as a metric for comparing the generalization performance of networks, without requiring labeled data. We find that sensitivity is decreased by applying popular methods which improve the generalization performance of the model, such as (1) using a deep network rather than a wide one, (2) adding convolutional layers to baseline classifiers instead of adding fully-connected layers, (3) using batch normalization, dropout and max-pooling, and (4) applying parameter initialization techniques.

This paper presents an experimental analysis about trade-offs in top-k classification accuracies on losses for deep leaning and proposal of a novel top-k loss. Commonly-used cross entropy (CE) is not guaranteed to optimize top-k prediction without infinite training data and model complexities. The objective is to clarify when CE sacrifices top-k accuracies to optimize top-1 prediction, and to design loss that improve top-k accuracy under such conditions. Our novel loss is basically CE modified by grouping temporal top-k classes as a single class. To obtain a robust decision boundary, we introduce an adaptive transition from normal CE to our loss, and thus call it top-k transition loss. It is demonstrated that CE is not always the best choice to learn top-k prediction in our experiments. First, we explore trade-offs between top-1 and top-k (=2) accuracies on synthetic datasets, and find a failure of CE in optimizing top-k prediction when we have complex data distribution for a given model to represent optimal top-1 prediction. Second, we compare top-k accuracies on CIFAR-100 dataset targeting top-5 prediction in deep learning. While CE performs the best in top-1 accuracy, in top-5 accuracy our loss performs better than CE except using one experimental setup. Moreover, our loss has been found to provide better top-k accuracies compared to CE at k larger than 10. As a result, a ResNet18 model trained with our loss reaches 99 % accuracy with k=25 candidates, which is a smaller candidate number than that of CE by 8.

We develop an approach to training deep networks while dynamically adjusting their architecture, driven by a principled combination of accuracy and sparsity objectives. Unlike conventional pruning approaches, our method adopts a gradual continuous relaxation of discrete network structure optimization and then samples sparse subnetworks, enabling efficient deep networks to be trained in a growing and pruning manner. Extensive experiments across CIFAR-10, ImageNet, PASCAL VOC, and Penn Treebank, with convolutional models for image classification and semantic segmentation, and recurrent models for language modeling, show that our training scheme yields efficient networks that are smaller and more accurate than those produced by competing pruning methods.

Deep Neural Networks (DNNs) are being used in various daily tasks such as object detection, speech processing, and machine translation. However, it is known that DNNs suffer from robustness problems -- perturbed inputs called adversarial samples leading to misbehaviors of DNNs. In this paper, we propose a black-box technique called Black-box Momentum Iterative Fast Gradient Sign Method (BMI-FGSM) to test the robustness of DNN models. The technique does not require any knowledge of the structure or weights of the target DNN. Compared to existing white-box testing techniques that require accessing model internal information such as gradients, our technique approximates gradients through Differential Evolution and uses approximated gradients to construct adversarial samples. Experimental results show that our technique can achieve 100% success in generating adversarial samples to trigger misclassification, and over 95% success in generating samples to trigger misclassification to a specific target output label. It also demonstrates better perturbation distance and better transferability. Compared to the state-of-the-art black-box technique, our technique is more efficient. Furthermore, we conduct testing on the commercial Aliyun API and successfully trigger its misbehavior within a limited number of queries, demonstrating the feasibility of real-world black-box attack.

The remarkable predictive performance of deep neural networks (DNNs) has led to their adoption in service domains of unprecedented scale and scope. However, the widespread adoption and growing commercialization of DNNs have underscored the importance of intellectual property (IP) protection. Devising techniques to ensure IP protection has become necessary due to the increasing trend of outsourcing the DNN computations on the untrusted accelerators in cloud-based services. The design methodologies and hyper-parameters of DNNs are crucial information, and leaking them may cause massive economic loss to the organization. Furthermore, the knowledge of DNN's architecture can increase the success probability of an adversarial attack where an adversary perturbs the inputs and alter the prediction. In this work, we devise a two-stage attack methodology "DeepPeep" which exploits the distinctive characteristics of design methodologies to reverse-engineer the architecture of building blocks in compact DNNs. We show the efficacy of "DeepPeep" on P100 and P4000 GPUs. Additionally, we propose intelligent design maneuvering strategies for thwarting IP theft through the DeepPeep attack and proposed "Secure MobileNet-V1". Interestingly, compared to vanilla MobileNet-V1, secure MobileNet-V1 provides a significant reduction in inference latency ($\approx$60%) and improvement in predictive performance ($\approx$2%) with very-low memory and computation overheads.

Program similarity is a fundamental concept, central to the solution of software engineering tasks such as software plagiarism, clone identification, code refactoring and code search. Accurate similarity estimation between programs requires an in-depth understanding of their structure, semantics and flow. A control flow graph (CFG), is a graphical representation of a program which captures its logical control flow and hence its semantics. A common approach is to estimate program similarity by analysing CFGs using graph similarity measures, e.g. graph edit distance (GED). However, graph edit distance is an NP-hard problem and computationally expensive, making the application of graph similarity techniques to complex software programs impractical. This study intends to examine the effectiveness of graph neural networks to estimate program similarity, by analysing the associated control flow graphs. We introduce funcGNN, which is a graph neural network trained on labeled CFG pairs to predict the GED between unseen program pairs by utilizing an effective embedding vector. To our knowledge, this is the first time graph neural networks have been applied on labeled CFGs for estimating the similarity between high-level language programs. Results: We demonstrate the effectiveness of funcGNN to estimate the GED between programs and our experimental analysis demonstrates how it achieves a lower error rate (0.00194), with faster (23 times faster than the quickest traditional GED approximation method) and better scalability compared with the state of the art methods. funcGNN posses the inductive learning ability to infer program structure and generalise to unseen programs. The graph embedding of a program proposed by our methodology could be applied to several related software engineering problems (such as code plagiarism and clone identification) thus opening multiple research directions.

Hyperparameters tuning is a time-consuming approach, particularly when the architecture of the neural network is decided as part of this process. For instance, in convolutional neural networks (CNNs), the selection of the number and the characteristics of the hidden (convolutional) layers may be decided. This implies that the search process involves the training of all these candidate network architectures. This paper describes a proposal to reuse the weights of hidden (convolutional) layers among different trainings to shorten this process. The rationale is that if a set of convolutional layers have been trained to solve a given problem, the weights calculated in this training may be useful when a new convolutional layer is added to the network architecture. This idea has been tested using the CIFAR-10 dataset, testing different CNNs architectures with up to 3 convolutional layers and up to 3 fully connected layers. The experiments compare the training time and the validation loss when reusing and not reusing convolutional layers. They confirm that this strategy reduces the training time while it even increases the accuracy of the resulting neural network. This finding opens up the future possibility of integrating this strategy in existing AutoML methods with the purpose of reducing the total search time.

Do you want to get a job at Google? If the answer is yes, these are the most important skills that will help you get an engineering job at Google, and also I will help you with how to gain these important skills. Many of us that have some sort of engineering background have a dream wish to work in a company like Google, which has a huge impact on our lives and will have a huge impact on our future. Google LLC is an American multinational technology company that specializes in Internet-related services and products, which include online advertising technologies, a search engine, cloud computing, software, and hardware. Being one of the biggest companies in the world, the skills requirements for a job position are quite a lot and require you to have at least a bachelor's degree in engineering.

Deep Learning is an extremely fast-moving field and the huge number of research papers and ideas can be overwhelming. Even seasoned researchers have a hard time telling company PR from real breakthroughs. The goal of this post is to review those ideas that have stood the test of time, which is perhaps the only significance test one should rely on. These ideas, or improvements of them, have been used over and over again. If you were to start in Deep Learning today, understanding and implementing each of these techniques would give you an excellent foundation for understanding recent research and working on your own projects.