It is common practice in deep learning to use overparameterized networks and train for as long as possible; there are numerous studies that show, both theoretically and empirically, that such practices surprisingly do not unduly harm the generalization performance of the classifier. In this paper, we empirically study this phenomenon in the setting of adversarially trained deep networks, which are trained to minimize the loss under worst-case adversarial perturbations. We find that overfitting to the training set does in fact harm robust performance to a very large degree in adversarially robust training across multiple datasets (SVHN, CIFAR-10, CIFAR-100, and ImageNet) and perturbation models ($\ell_\infty$ and $\ell_2$). Based upon this observed effect, we show that the performance gains of virtually all recent algorithmic improvements upon adversarial training can be matched by simply using early stopping. We also show that effects such as the double descent curve do still occur in adversarially trained models, yet fail to explain the observed overfitting. Finally, we study several classical and modern deep learning remedies for overfitting, including regularization and data augmentation, and find that no approach in isolation improves significantly upon the gains achieved by early stopping. All code for reproducing the experiments as well as pretrained model weights and training logs can be found at https://github.com/locuslab/robust_overfitting.

ML models are increasingly deployed in settings with real world interactions such as vehicles, but unfortunately, these models can fail in systematic ways. To prevent errors, ML engineering teams monitor and continuously improve these models. We propose a new abstraction, model assertions, that adapts the classical use of program assertions as a way to monitor and improve ML models. Model assertions are arbitrary functions over a model's input and output that indicate when errors may be occurring, e.g., a function that triggers if an object rapidly changes its class in a video. We propose methods of using model assertions at all stages of ML system deployment, including runtime monitoring, validating labels, and continuously improving ML models. For runtime monitoring, we show that model assertions can find high confidence errors, where a model returns the wrong output with high confidence, which uncertainty-based monitoring techniques would not detect. For training, we propose two methods of using model assertions. First, we propose a bandit-based active learning algorithm that can sample from data flagged by assertions and show that it can reduce labeling costs by up to 40% over traditional uncertainty-based methods. Second, we propose an API for generating "consistency assertions" (e.g., the class change example) and weak labels for inputs where the consistency assertions fail, and show that these weak labels can improve relative model quality by up to 46%. We evaluate model assertions on four real-world tasks with video, LIDAR, and ECG data.

Natural Language Processing (NLP) helps empower intelligent machines by enhancing a better understanding of the human language for linguistic-based human-computer communication. Recent developments in computational power and the advent of large amounts of linguistic data have heightened the need and demand for automating semantic analysis using data-driven approaches. The utilization of data-driven strategies is pervasive now due to the significant improvements demonstrated through the usage of deep learning methods in areas such as Computer Vision, Automatic Speech Recognition, and in particular, NLP. This survey categorizes and addresses the different aspects and applications of NLP that have benefited from deep learning. It covers core NLP tasks and applications and describes how deep learning methods and models advance these areas. We further analyze and compare different approaches and state-of-the-art models.

Intrusion detection system (IDS) plays an essential role in computer networks protecting computing resources and data from outside attacks. Recent IDS faces challenges improving flexibility and efficiency of the IDS for unexpected and unpredictable attacks. Deep neural network (DNN) is considered popularly for complex systems to abstract features and learn as a machine learning technique. In this paper, we propose a deep learning approach for developing the efficient and flexible IDS using one-dimensional Convolutional Neural Network (1D-CNN). Two-dimensional CNN methods have shown remarkable performance in detecting objects of images in computer vision area. Meanwhile, the 1D-CNN can be used for supervised learning on time-series data. We establish a machine learning model based on the 1D-CNN by serializing Transmission Control Protocol/Internet Protocol (TCP/IP) packets in a predetermined time range as an invasion Internet traffic model for the IDS, where normal and abnormal network traffics are categorized and labeled for supervised learning in the 1D-CNN. We evaluated our model on UNSW\_NB15 IDS dataset to show the effectiveness of our method. For comparison study in performance, machine learning-based Random Forest (RF) and Support Vector Machine (SVM) models in addition to the 1D-CNN with various network parameters and architecture are exploited. In each experiment, the models are run up to 200 epochs with a learning rate in 0.0001 on imbalanced and balanced data. 1D-CNN and its variant architectures have outperformed compared to the classical machine learning classifiers. This is mainly due to the reason that CNN has the capability to extract high-level feature representations that represent the abstract form of low-level feature sets of network traffic connections.

Neural-symbolic computing has now become the subject of interest of both academic and industry research laboratories. Graph Neural Networks (GNN) have been widely used in relational and symbolic domains, with widespread application of GNNs in combinatorial optimization, constraint satisfaction, relational reasoning and other scientific domains. The need for improved explainability, interpretability and trust of AI systems in general demands principled methodologies, as suggested by neural-symbolic computing. In this paper, we review the state-of-the-art on the use of GNNs as a model of neural-symbolic computing. This includes the application of GNNs in several domains as well as its relationship to current developments in neural-symbolic computing.

For centuries, flow visualization has been the art of making fluid motion visible in physical and biological systems. Although such flow patterns can be, in principle, described by the Navier-Stokes equations, extracting the velocity and pressure fields directly from the images is challenging. We addressed this problem by developing hidden fluid mechanics (HFM), a physics-informed deep-learning framework capable of encoding the Navier-Stokes equations into the neural networks while being agnostic to the geometry or the initial and boundary conditions. We demonstrate HFM for several physical and biomedical problems by extracting quantitative information for which direct measurements may not be possible. HFM is robust to low resolution and substantial noise in the observation data, which is important for potential applications.

Amazon has announced it has a new artificial intelligence (AI) model that helps convert text to images to aid in searching for products, according to a blog post by the company. "Generative adversarial networks (GANs), which were first introduced in 2014, have proven remarkably successful at generating synthetic images. A GAN consists of two networks, one that tries to produce convincing fakes, and one that tries to distinguish fakes from real examples. The two networks are trained together, and the competition between them can converge quickly on a useful generative model," the post said. Someone who was searching for "women's black pants" could type that in to get an image, but then when they added more words, like "capri" or "petite," new images would show up as well as old ones.

The first 2 courses are offered in Python and C while the Deep Learning with PyTorch course is in Python. The content will have a mix of theory and code with explanations through both Text and Video. Apart from this, there will be quizzes, assignments and projects. Each of them will be of varying difficulty and students can also choose which assignment or project they want to work on. Every assignment/project has some points and we take the best "x" out of "y" assignments/projects.

All email providers today offer a special mail category known as'Spam', but how does the email program figure out the criteria for this type of categorization? The answer lies in Machine Learning. It could also be based on certain specific/typical patterns of email releases or roll outs. So when you receive a spam email, machine learning algorithms at the backend enable the system to identify and sort such messages automatically, placing it in the Spam folder and saving you the annoyance of doing it yourself while keeping you safe online. I often receive emails that claim I have won the lottery.

The Machine Learning team at Mozilla continues work on DeepSpeech, an automatic speech recognition (ASR) engine which aims to make speech recognition technology and trained models openly available to developers. DeepSpeech is a deep learning-based ASR engine with a simple API. We also provide pre-trained English models. Our latest release, version v0.6, offers the highest quality, most feature-packed model so far. In this overview, we'll show how DeepSpeech can transform your applications by enabling client-side, low-latency, and privacy-preserving speech recognition capabilities.