One of the challenges in analyzing learning algorithms is the circular entanglement between the objective value and the stochastic noise. This is also known as the "chicken and egg" phenomenon and traditionally, there is no principled way to tackle this issue. People solve the problem by utilizing the special structure of the dynamic, and hence the analysis would be difficult to generalize. In this work, we present a streamlined three-step recipe to tackle the "chicken and egg" problem and give a general framework for analyzing stochastic dynamics in learning algorithms. Our framework composes standard techniques from probability theory, such as stopping time and martingale concentration. We demonstrate the power and flexibility of our framework by giving a unifying analysis for three very different learning problems with the last iterate and the strong uniform high probability convergence guarantee. The problems are stochastic gradient descent for strongly convex functions, streaming principal component analysis, and linear bandit with stochastic gradient descent updates. We either improve or match the state-of-the-art bounds on all three dynamics.

Dog's minds are being read! Researchers have used fMRI (functional magnetic resonance imaging) scans of dogs' brains and a machine learning tool to reconstruct what the pooch is seeing. The results suggest that dogs are more interested in what is happening than who or what is involved. The results of the experiment conducted at Emory University in Georgia in the US are published in the Journal of Visualized Experiments. Two unrestrained dogs were shown three 30-minute videos.

An innovative chatbot designed for sharing critical information about sexual and reproductive health (SRH) with young people in India is demonstrating how artificial intelligence (AI) applications can engage vulnerable and hard-to-reach population segments. Working with the Population Foundation of India (PFI), Helen Wang, associate professor of communication, College of Arts and Sciences, examined the user-centered design and engagement of SnehAI, the first Hinglish (Hindi and English) chatbot purposefully developed for social and behavioral change. "Many AI technologies today are motivated by profit, but we must also be aware that AI can be leveraged in ways that facilitate social and behavior change," says Wang, who specializes in entertainment-education and storytelling as instruments for health promotion. "SnehAI is a powerful testimonial of the vital potential that lies in AI for good." The findings from Wang's instrumental case study appear in the Journal of Medical Internet Research.

A group of artificial intelligence graduate students from several prestigious international universities concluded their participation in the second edition of the Global AI Summit, which concluded last week in Riyadh. The students also visited Masmak Palace in the center of Riyadh to be briefed on the history of the capital. The students represented six countries, joined by several Saudi scholarship students in the same specialization. Their participation came within the knowledge exchange initiative launched by the Saudi Data and Artificial Intelligence Authority, which hosted 19 male and female students of different nationalities including the US, the UK, India, Jordan, Algeria, South Korea and Nigeria. These students study at international universities and institutes, including the Sorbonne University in Paris, Oxford University, University College London, Durham University, Nottingham University, Sussex University in the UK, the Massachusetts Institute of Technology in the US, and King's College London.

Hewlett Packard Enterprise announced that King Abdullah University of Science and Technology (KAUST) selected HPE to build its next-generation supercomputer, Shaheen III, to deliver state-of-the-art supercomputing and artificial intelligence (AI) capabilities for advancing research in fields such as food, water, energy and the environment. "Powered by AMD EPYC processors, Shaheen III will enable new discoveries that will have regional and global impacts across climate, clean energy and tectonic plate modeling, all made possible by the collaboration between KAUST scientists and HPE." Supercomputing capacity has become increasingly vital to global innovation, industry competitiveness and economic growth. From accelerating vaccine discovery to fight a pandemic, advancing clean energy systems to increase sustainability, to enabling new possibilities in AI, supercomputing is a core technology to solving the world's most challenging scientific and engineering problems. Shaheen III, set to be 20 times faster than KAUST's existing system, will be the most powerful supercomputer in the Middle East to address critical areas that have a societal and environmental impact. Built by HPE, the world's leading supercomputer provider, the new Shaheen III system will revolutionize KAUST's ability to process vast amounts of data at immense speed and scale, enabling its users to unlock discoveries that it could not have before, and realize new potentials for AI.

Real-time interception of a fast-moving object by a robotic arm in cluttered environments filled with static or dynamic obstacles permits only tens of milliseconds for reaction times, hence quite challenging and arduous for state-of-the-art robotic planning algorithms to perform multiple robotic skills, for instance, catching the dynamic object and avoiding obstacles, in parallel. This paper proposes an unified framework of robotic path planning through embedding the high-dimensional temporal information contained in the event stream to distinguish between safe and colliding trajectories into a low-dimension space manifested with a pre-constructed 2D densely connected graph. We then leverage a fast graph-traversing strategy to generate the motor commands necessary to effectively avoid the approaching obstacles while simultaneously intercepting a fast-moving objects. The most distinctive feature of our methodology is to conduct both object interception and obstacle avoidance within the same algorithm framework based on deep manifold learning. By leveraging a highly efficient diffusion-map based variational autoencoding and Extended Kalman Filter(EKF), we demonstrate the effectiveness of our approach on an autonomous 7-DoF robotic arm using only onboard sensing and computation. Our robotic manipulator was capable of avoiding multiple obstacles of different sizes and shapes while successfully capturing a fast-moving soft ball thrown by hand at normal speed in different angles. Complete video demonstrations of our experiments can be found in https://sites.google.com/view/multirobotskill/home.

Vocal bursts play an important role in communicating affect, making them valuable for improving speech emotion recognition. Here, we present our approach for classifying vocal bursts and predicting their emotional significance in the ACII Affective Vocal Burst Workshop & Challenge 2022 (A-VB). We use a large self-supervised audio model as shared feature extractor and compare multiple architectures built on classifier chains and attention networks, combined with uncertainty loss weighting strategies. Our approach surpasses the challenge baseline by a wide margin on all four tasks.

Addressing a real world sequential decision problem with Reinforcement Learning (RL) usually starts with the use of a simulated environment that mimics real conditions. We present a novel open source RL environment for realistic crop management tasks. gym-DSSAT is a gym interface to the Decision Support System for Agrotechnology Transfer (DSSAT), a high fidelity crop simulator. DSSAT has been developped over the last 30 years and is widely recognized by agronomists. gym-DSSAT comes with predefined simulations based on real world maize experiments. The environment is as easy to use as any gym environment. We provide performance baselines using basic RL algorithms. We also briefly outline how the monolithic DSSAT simulator written in Fortran has been turned into a Python RL environment. Our methodology is generic and may be applied to similar simulators. We report on very preliminary experimental results which suggest that RL can help researchers to improve sustainability of fertilization and irrigation practices.

The application of models to assess the risk of the physical impacts of weather and climate and their subsequent consequences for society and business is of the utmost importance in our changing climate. The operation of such models is historically bespoke and constrained to specific compute infrastructure, driving datasets and predefined configurations. These constraints introduce challenges with scaling model runs and putting the models in the hands of interested users. Here we present a cloud-based modular framework for the deployment and operation of geospatial models, initially applied to climate impacts. The Climate Impact Modelling Frameworks (CIMF) enables the deployment of modular workflows in a dynamic and flexible manner. Users can specify workflow components in a streamlined manner, these components can then be easily organised into different configurations to assess risk in different ways and at different scales. This also enables different models (physical simulation or machine learning models) and workflows to be connected to produce combined risk assessment. Flood modelling is used as an end-to-end example to demonstrate the operation of CIMF.

In this paper, we develop a neural network model to predict future human motion from an observed human motion history. We propose a non-autoregressive transformer architecture to leverage its parallel nature for easier training and fast, accurate predictions at test time. The proposed architecture divides human motion prediction into two parts: 1) the human trajectory, which is the hip joint 3D position over time and 2) the human pose which is the all other joints 3D positions over time with respect to a fixed hip joint. We propose to make the two predictions simultaneously, as the shared representation can improve the model performance. Therefore, the model consists of two sets of encoders and decoders. First, a multi-head attention module applied to encoder outputs improves human trajectory. Second, another multi-head self-attention module applied to encoder outputs concatenated with decoder outputs facilitates learning of temporal dependencies. Our model is well-suited for robotic applications in terms of test accuracy and speed, and compares favorably with respect to state-of-the-art methods. We demonstrate the real-world applicability of our work via the Robot Follow-Ahead task, a challenging yet practical case study for our proposed model.