One of the biggest machine learning trends you can expect to see in 2021 and beyond is the broader adoption of a machine learning segment called deep learning and a method it employs called weak supervision. This method is faster, more streamlined, and has many benefits. Moving forward, companies will assess how they can leverage this in their business to execute on desired automation tasks and learn and predict the best ways to accomplish task completion– all without the heavy burden of human intervention. Weak supervision is bringing us ever closer to software that can think and act on its own, and here we will break down the pitfalls of the past, the methods of the future, and what it means for your business. Further solidifying weak supervision as an entirely accepted practice in our immediate future is the added element of Explainable AI (XAI) that serves to future-proof this new approach to ramping up an AI system.

A new deep learning model can be a solution for diabetic retinopathy. Sankara Eye Foundation and Leben Care, a Singapore-based company are collaborating to provide diabetic retinopathy relief in India. The companies are working on an Intel-powered, cloud-based artificial intelligence solution that uses deep learning to detect retinal conditions within a short span. Called Netra.AI, the accuracy of this system matches that of human doctors and can drastically reduce screening burdens. Prakash Mallya, vice president and managing director of sales, marketing, and communications group at Intel India says, "The use of AI to improve disease detection and prevention is a critical step for the healthcare industry and a giant leap for humankind. India has one of the largest diabetic populations in the world and diabetic retinopathy is the major cause of vision loss and blindness in persons of working age. WIth Netra.AI, Sankara Eye Foundation, and Leben Care have leveraged the power of Intel Xeon Scalable processors and built-in Intel Deep Learning (DL) Boost to accurately detect DR and enable timely treatment to effectively combat avoidable vision impairment and blindness in diabetic patients."

Researchers from MIT have developed a new AI approach that could soon find its way into self-driving cars and industrial robots in smart factories. Designed to handle unpredictable interactions safely, the deep-learning algorithm promises to enhance the robustness of AI systems in safety-critical scenarios. From avoiding a pedestrian dashing across the road in unusually bad weather to overcoming the malicious obstruction of sensors in a manufacturing plant, the new system can enable AI systems to react in a robust manner even when critical inputs deviate due to either unreliable inputs or noise. The details of this new approach are outlined in a study by Michael Everett, Björn Lütjens, and Jonathan How from MIT. Titled "Certifiable robustness to adversarial state uncertainty in deep reinforcement learning", the study was published last month in IEEE's Transactions on Neural Networks and Learning Systems. The algorithm works by building a healthy "skepticism" of the measurements and inputs AI systems receive to help machines to navigate our real, imperfect world.

Practically, deep learning is just a subset of machine learning, while ML is Artificial Intelligence itself. However, its capabilities are different.

Despite recent research efforts, the vision of automatic code generation through API recommendation has not been realized. Accuracy and expressiveness challenges of API recommendation needs to be systematically addressed. We present a new neural network-based approach, Multi-HyLSTM for API recommendation --targeting cryptography-related code. Multi-HyLSTM leverages program analysis to guide the API embedding and recommendation. By analyzing the data dependence paths of API methods, we train embedding and specialize a multi-path neural network architecture for API recommendation tasks that accurately predict the next API method call. We address two previously unreported programming language-specific challenges, differentiating functionally similar APIs and capturing low-frequency long-range influences. Our results confirm the effectiveness of our design choices, including program-analysis-guided embedding, multi-path code suggestion architecture, and low-frequency long-range-enhanced sequence learning, with high accuracy on top-1 recommendations. We achieve a top-1 accuracy of 91.41% compared with 77.44% from the state-of-the-art tool SLANG. In an analysis of 245 test cases, compared with the commercial tool Codota, we achieve a top-1 recommendation accuracy of 88.98%, which is significantly better than Codota's accuracy of 64.90%. We publish our data and code as a large Java cryptographic code dataset.

Lack of adequate training samples and noisy high-dimensional features are key challenges faced by Motor Imagery (MI) decoding algorithms for electroencephalogram (EEG) based Brain-Computer Interface (BCI). To address these challenges, inspired from neuro-physiological signatures of MI, this paper proposes a novel Filter-Bank Convolutional Network (FBCNet) for MI classification. FBCNet employs a multi-view data representation followed by spatial filtering to extract spectro-spatially discriminative features. This multistage approach enables efficient training of the network even when limited training data is available. More significantly, in FBCNet, we propose a novel Variance layer that effectively aggregates the EEG time-domain information. With this design, we compare FBCNet with state-of-the-art (SOTA) BCI algorithm on four MI datasets: The BCI competition IV dataset 2a (BCIC-IV-2a), the OpenBMI dataset, and two large datasets from chronic stroke patients. The results show that, by achieving 76.20% 4-class classification accuracy, FBCNet sets a new SOTA for BCIC-IV-2a dataset. On the other three datasets, FBCNet yields up to 8% higher binary classification accuracies. Additionally, using explainable AI techniques we present one of the first reports about the differences in discriminative EEG features between healthy subjects and stroke patients. Also, the FBCNet source code is available at https://github.com/ravikiran-mane/FBCNet.

Over recent years, a myriad of novel convolutional network architectures have been developed to advance state-of-the-art performance on challenging recognition tasks. As computational resources improve, a great deal of effort has been placed in efficiently scaling up existing designs and generating new architectures with Neural Architecture Search (NAS) algorithms. While network topology has proven to be a critical factor for model performance, we show that significant gains are being left on the table by keeping topology static at inference-time. Due to challenges such as scale variation, we should not expect static models configured to perform well across a training dataset to be optimally configured to handle all test data. In this work, we seek to expose the exciting potential of inference-time-dynamic models. By allowing just four layers to dynamically change configuration at inference-time, we show that existing off-the-shelf models like ResNet-50 are capable of over 95% accuracy on ImageNet. This level of performance currently exceeds that of models with over 20x more parameters and significantly more complex training procedures.

Adversarial training enhances the robustness of deep neural networks at the cost of decreased accuracy on the clean test samples [Goodfellow et al., 2014, Madry et al., 2017, Sinha et al., 2017]. Though the model can fit the training data perfectly in adversarial training, the generalization error on clean test dataset increases compared with non-adversarially trained models. For example, in the rightmost panel of Figure 1(b), we can see that, even if an adversarially trained model achieves almost zero error on the clean training data (up to a certain level of perturbation ε), the error on the clean test data (the blue curve) keeps increasing with ε. Hence to improve both robustness and accuracy of (adversarially trained) deep networks, it is crucial to understand the cause for this increased "generalization gap" between errors on the (clean) training dataset and (clean) test dataset. In this work, to better understand the generalization gap, we turn to a standard tool of statistical learning theory, the bias-variance decomposition [Markov, 1900, Lehmann, 1983, Casella and Berger, 1990, Hastie et al., 2009, Geman et al., 1992]. A large variance corresponds to the instability of the model, whereas a large bias suggests that the model predicts poorly on average. Bias and variance provide more information about the generalization gap than just test error alone: We can better understand whether an explanation works by checking whether it predicts both the bias and variance. How does adversarial training affect the bias and the variance?

The problem of classifying high-dimensional shapes in real-world data grows in complexity as the dimension of the space increases. For the case of identifying convex shapes of different geometries, a new classification framework has recently been proposed in which the intersections of a set of one-dimensional representations, called rays, with the boundaries of the shape are used to identify the specific geometry. This ray-based classification (RBC) has been empirically verified using a synthetic dataset of two- and three-dimensional shapes [1] and, more recently, has also been validated experimentally [2]. Here, we establish a bound on the number of rays necessary for shape classification, defined by key angular metrics, for arbitrary convex shapes. For two dimensions, we derive a lower bound on the number of rays in terms of the shape's length, diameter, and exterior angles. For convex polytopes in R^N, we generalize this result to a similar bound given as a function of the dihedral angle and the geometrical parameters of polygonal faces. This result enables a different approach for estimating high-dimensional shapes using substantially fewer data elements than volumetric or surface-based approaches.

A potential Mars Sample Return (MSR) architecture is being jointly studied by NASA and ESA. As currently envisioned, the MSR campaign consists of a series of 3 missions: sample cache, fetch and return to Earth. In this paper, we focus on the fetch part of the MSR, and more specifically the problem of autonomously detecting and localizing sample tubes deposited on the Martian surface. Towards this end, we study two machine-vision based approaches: First, a geometry-driven approach based on template matching that uses hard-coded filters and a 3D shape model of the tube; and second, a data-driven approach based on convolutional neural networks (CNNs) and learned features. Furthermore, we present a large benchmark dataset of sample-tube images, collected in representative outdoor environments and annotated with ground truth segmentation masks and locations. The dataset was acquired systematically across different terrain, illumination conditions and dust-coverage; and benchmarking was performed to study the feasibility of each approach, their relative strengths and weaknesses, and robustness in the presence of adverse environmental conditions.