The MineRL BASALT competition has served to catalyze advances in learning from human feedback through four hard-to-specify tasks in Minecraft, such as create and photograph a waterfall. Given the completion of two years of BASALT competitions, we offer to the community a formalized benchmark through the BASALT Evaluation and Demonstrations Dataset (BEDD), which serves as a resource for algorithm development and performance assessment. BEDD consists of a collection of 26 million image-action pairs from nearly 14,000 videos of human players completing the BASALT tasks in Minecraft. It also includes over 3,000 dense pairwise human evaluations of human and algorithmic agents. These comparisons serve as a fixed, preliminary leaderboard for evaluating newly-developed algorithms. To enable this comparison, we present a streamlined codebase for benchmarking new algorithms against the leaderboard. In addition to presenting these datasets, we conduct a detailed analysis of the data from both datasets to guide algorithm development and evaluation.

Even when massively overparameterized, deep neural networks show a remarkable ability to generalize. Research on this phenomenon has focused on generalization within distribution, via smooth interpolation. Yet in some settings neural networks also learn to extrapolate to data far beyond the bounds of the original training set, sometimes even allowing for infinite generalization, implying that an algorithm capable of solving the task has been learned. Here we undertake a case study of the learning dynamics of recurrent neural networks (RNNs) trained on the streaming parity task in order to develop an effective theory of algorithm development. The streaming parity task is a simple but nonlinear task defined on sequences up to arbitrary length. We show that, with sufficient finite training experience, RNNs exhibit a phase transition to perfect infinite generalization. Using an effective theory for the representational dynamics, we find an implicit representational merger effect which can be interpreted as the construction of a finite automaton that reproduces the task. Overall, our results disclose one mechanism by which neural networks can generalize infinitely from finite training experience.

The MineRL BASALT competition has served to catalyze advances in learning from human feedback through four hard-to-specify tasks in Minecraft, such as create and photograph a waterfall. Given the completion of two years of BASALT competitions, we offer to the community a formalized benchmark through the BASALT Evaluation and Demonstrations Dataset (BEDD), which serves as a resource for algorithm development and performance assessment. BEDD consists of a collection of 26 million image-action pairs from nearly 14,000 videos of human players completing the BASALT tasks in Minecraft. It also includes over 3,000 dense pairwise human evaluations of human and algorithmic agents. These comparisons serve as a fixed, preliminary leaderboard for evaluating newly-developed algorithms.

This paper presents a unique outdoor aerial visual-inertial-LiDAR dataset captured using a multi-sensor payload to promote the global navigation satellite system (GNSS)-denied navigation research. The dataset features flight distances ranging from 300m to 5km, collected using a DJI M600 hexacopter drone and the National Research Council (NRC) Bell 412 Advanced Systems Research Aircraft (ASRA). The dataset consists of hardware synchronized monocular images, IMU measurements, 3D LiDAR point-clouds, and high-precision real-time kinematic (RTK)-GNSS based ground truth. Ten datasets were collected as ROS bags over 100 mins of outdoor environment footage ranging from urban areas, highways, hillsides, prairies, and waterfronts. The datasets were collected to facilitate the development of visual-inertial-LiDAR odometry and mapping algorithms, visual-inertial navigation algorithms, object detection, segmentation, and landing zone detection algorithms based upon real-world drone and full-scale helicopter data. All the datasets contain raw sensor measurements, hardware timestamps, and spatio-temporally aligned ground truth. The intrinsic and extrinsic calibrations of the sensors are also provided along with raw calibration datasets. A performance summary of state-of-the-art methods applied on the datasets is also provided.

Increasing attention is being paid to millimeter-wave (mmWave), 30 GHz to 300 GHz, and terahertz (THz), 300 GHz to 10 THz, sensing applications including security sensing, industrial packaging, medical imaging, and non-destructive testing. Traditional methods for perception and imaging are challenged by novel data-driven algorithms that offer improved resolution, localization, and detection rates. Over the past decade, deep learning technology has garnered substantial popularity, particularly in perception and computer vision applications. Whereas conventional signal processing techniques are more easily generalized to various applications, hybrid approaches where signal processing and learning-based algorithms are interleaved pose a promising compromise between performance and generalizability. Furthermore, such hybrid algorithms improve model training by leveraging the known characteristics of radio frequency (RF) waveforms, thus yielding more efficiently trained deep learning algorithms and offering higher performance than conventional methods. This dissertation introduces novel hybrid-learning algorithms for improved mmWave imaging systems applicable to a host of problems in perception and sensing. Various problem spaces are explored, including static and dynamic gesture classification; precise hand localization for human computer interaction; high-resolution near-field mmWave imaging using forward synthetic aperture radar (SAR); SAR under irregular scanning geometries; mmWave image super-resolution using deep neural network (DNN) and Vision Transformer (ViT) architectures; and data-level multiband radar fusion using a novel hybrid-learning architecture. Furthermore, we introduce several novel approaches for deep learning model training and dataset synthesis.

The entry barrier to the world of algorithms is getting lower by the day. That means anybody with the right goal and skills can find out great algorithms for Machine Learning (ML) and Artificial Intelligence (AI) tasks - computer vision, natural language processing, recommendation systems, or even autonomous driving. Open-source computing has come a long way and plenty of open-source initiatives are propelling the vehicles of data science, digital analytics, and ML. Researchers around the universities and corporate R&D labs are creating new algorithms and ML techniques every day. We can safely say that algorithms, programming frameworks, ML packages, and even tutorials and courses on how to learn these techniques are no longer scarce resources.

The nuclear medicine field has seen a rapid expansion of academic and commercial interest in developing artificial intelligence (AI) algorithms. Users and developers can avoid some of the pitfalls of AI by recognizing and following best practices in AI algorithm development. In this article, recommendations on technical best practices for developing AI algorithms in nuclear medicine are provided, beginning with general recommendations and then continuing with descriptions of how one might practice these principles for specific topics within nuclear medicine. This report was produced by the AI Task Force of the Society of Nuclear Medicine and Molecular Imaging.

The entry barrier to the world of algorithms is getting lower by the day. That means anybody with the right goal and skills can find out great algorithms for Machine Learning (ML) and Artificial Intelligence (AI) tasks - computer vision, natural language processing, recommendation systems, or even autonomous driving. Open-source computing has come a long way and plenty of open-source initiatives are propelling the vehicles of data science, digital analytics, and ML. Researchers around the universities and corporate R&D labs are creating new algorithms and ML techniques every day. We can safely say that algorithms, programming frameworks, ML packages, and even tutorials and courses on how to learn these techniques are no longer scarce resources.

From self-driving cars to virtual travel agents, artificial intelligence has quickly transformed the landscape for nearly every industry. The technology is also employed in healthcare to help with clinical decision support, imaging and triage. However, using AI in a healthcare setting poses a unique set of ethical and logistical challenges. MobiHealthNews asked health tech vet Muhammad Babur, a program manager at the Mayo Clinic, about the potential challenges and ethics behind using AI in healthcare ahead of his upcoming discussion at HIMSS22. MobiHealthNews: What are some of the challenges to using AI in healthcare? Babur: The challenges that we face in healthcare are unique and more consequential.

A recent survey of over 225 enterprise Data Scientists, AI technologists and business stakeholders involved in active AI and machine learning (ML) projects, suggests that for most organizations, it's still early days for AI technology. The AI market is projected to become a $190 billion industry by 2025 ( according to Markets and Markets), and global spending on cognitive and AI systems is expected to reach $35.8 billion in 2029, an increase of 44.0% over the amount spent in 2018 (according to IDC). This research suggests AI is advanced and on the move, already being undertaken by large enterprises and ready to make an impact on how we live and work. But it is still early days for AI when it comes to the implementation of AI in organisations and there are reasons for that. An AI system requires meticulous training before it can perform its intended function. When that function involves something as complex as making human-like judgments about images or videos – "seeing," in other words – the system must be exposed to enormous volumes of accurately labeled and annotated training data.