I highly recommend reading the McKinsey Global Institute's new report, "Reskilling China: Transforming The World's Largest Workforce Into Lifelong Learners", which focuses on the country's biggest employment challenge, re-training its workforce and the adoption of practices such as lifelong learning to address the growing digital transformation of its productive fabric. How to transform the country that has become the factory of the world, where manual assembly was the cheapest due to its low labor costs, into an artificial intelligence giant, with the largest public blockchain infrastructure in the world, a digital currency in an advanced stage of development that will see an end to cash payments, along with the world's largest 5G network? Xi Jinping's state capitalism is transforming the Asian giant: the possibility of drawing up and maintaining long-term strategies thanks to political stability is driving change at an unprecedented rate, that includes autonomous driving and digital healthcare, advanced retail or even livestock farming. No matter where you look: the modernization and robotization of Chinese assembly factories has led to enormous reductions in the size of their workforces, which, moreover, immediately correspond not only to an increase in their production capacity, but also to a drastic reduction in the number of errors. And the COVID-19 pandemic, far from slowing the process, has accelerated it even further.

The program aims to prepare community college students for careers tapping AI skills. Intel said Tuesday it's expanding a program that aims to educate tomorrow's engineers and technologists on the intricacies of artificial intelligence and help them find jobs in their chosen field. The AI for Workforce Program offers students courses on data collection, computer vision, AI model training, coding, the societal impacts and ethics of AI technology. Students who complete the program will be awarded a certificate or associate degree in artificial intelligence. The program began as a collaboration with an Arizona community college but is being expanded to 18 community colleges in 11 states through a partnership with Dell Technologies, which will provide guidance on how best to configure AI labs for teaching in-person, hybrid and online students.

Due to changes in model dynamics or unexpected disturbances, an autonomous robotic system may experience unforeseen challenges during real-world operations which may affect its safety and intended behavior: in particular actuator and system failures and external disturbances are among the most common causes of degraded mode of operation. To deal with this problem, in this work, we present a meta-learning-based approach to improve the trajectory tracking performance of an unmanned aerial vehicle (UAV) under actuator faults and disturbances which have not been previously experienced. Our approach leverages meta-learning to train a model that is easily adaptable at runtime to make accurate predictions about the system's future state. A runtime monitoring and validation technique is proposed to decide when the system needs to adapt its model by considering a data pruning procedure for efficient learning. Finally, the reference trajectory is adapted based on future predictions by borrowing feedback control logic to make the system track the original and desired path without needing to access the system's controller. The proposed framework is applied and validated in both simulations and experiments on a faulty UAV navigation case study demonstrating a drastic increase in tracking performance.

Seeking to represent natural language, researchers have found language models (LM) with Sesame Street-inspired names [1] [2] [3] to be incredibly effective methods of producing language representations (LR). These LM's have leverage transfer learning by training on a large text corpus to learn a good representation of language which can then be used in a down steam task like Question Answering or Entity Resolution. While these LMs have shown to be excellent methods to enable language understanding, the ability to train these models is becoming increasingly computationally expensive [4]. Since model performance is closely tied to the size of training data, model size, and compute used to train [5] the bulk of existing research has focused on scaling these aspects without much focus on increasing efficiency of training. Seeking to explore what methods could be used to make LM training more efficient we study the effect of curriculum learning by training ELMo with a wide variety of curricula.

To increase autonomy in reinforcement learning, agents need to learn useful behaviours without reliance on manually designed reward functions. To that end, skill discovery methods have been used to learn the intrinsic options available to an agent using task-agnostic objectives. However, without the guidance of task-specific rewards, emergent behaviours are generally useless due to the under-constrained problem of skill discovery in complex and high-dimensional spaces. This paper proposes a framework for guiding the skill discovery towards the subset of expert-visited states using a learned state projection. We apply our method in various reinforcement learning (RL) tasks and show that such a projection results in more useful behaviours.

Following an online pilot in the fall of 2020 with Maricopa County Community College District, Intel is expanding its AI for Workforce Program to include 18 additional schools in 11 states, including California, New Mexico and Michigan. With the expansion, more than 800,000 students can take part in a curriculum designed by the company, at the end of which they can earn a certificate or associate degree in artificial intelligence. The program includes courses on data collection, computer vision, model training, coding and AI ethics. In addition to designing the curriculum, Intel has provided training and technical advice to the college faculty involved in the program. Dell is also helping with technical and infrastructure expertise.

The transformer architecture has shown an uncanny ability to model not only language but also images and proteins. New research found that it can apply what it learns from the first domain to the others. What's new: Kevin Lu and colleagues at UC Berkeley, Facebook, and Google devised Frozen Pretrained Transformer (FPT). After pretraining a transformer network on language data, they showed that it could perform vision, mathematical, and logical tasks without fine-tuning its core layers. Key insight: Transformers pick up on patterns in an input sequence, be it words in a novel, pixels in an image, or amino acids in a protein.

You probably interact with computers daily, but let's be specific about what we mean when we talk about computers and programming. Programming tells the computer how to receive, process, and then store this data. When someone writes a program, that person gives the computer a set of commands to follow. Programming, at its core, takes a big problem and breaks it down into smaller and smaller problems until they are small enough that we can tell the computer to solve the problem. We are going to discuss what programming languages are, what are the main differences, and where you can learn them.

Here are the results of the KDnuggets Poll inspired by this blog: Relax! As advances in AI continue to progress in leaps and bounds, accessibility to data science at a base level has become increasingly democratized. Traditional entry barriers to the field such as a lack of data and computing power have been swept aside with a continuous supply of new data startups popping up(some offering access for as little as a cup of coffee a day) and all powerful cloud computing removing the need for expensive onsite hardware. Rounding out the trinity of prerequisites, is the skill and know-how to implement, which has arguably become the most ubiquitous aspect of data science. One does not need to look far to find online tutorials touting taglines like "implement X model in seconds", "apply Z method to your data in just a few lines of code". In a digital world, instant gratification has become the name of the game.

Background: The objective of this study was to characterize patients with hyponatremia at hospital admission into clusters using an unsupervised machine learning approach, and to evaluate the short- and long-term mortality risk among these distinct clusters. Methods: We performed consensus cluster analysis based on demographic information, principal diagnoses, comorbidities, and laboratory data among 11,099 hospitalized adult hyponatremia patients with an admission serum sodium below 135 mEq/L. The standardized mean difference was utilized to identify each cluster’s key features. We assessed the association of each hyponatremia cluster with hospital and one-year mortality using logistic and Cox proportional hazard analysis, respectively. Results: There were three distinct clusters of hyponatremia patients: 2033 (18%) in cluster 1, 3064 (28%) in cluster 2, and 6002 (54%) in cluster 3. Among these three distinct clusters, clusters 3 patients were the youngest, had lowest comorbidity burden, and highest kidney function. Cluster 1 patients were more likely to be admitted for genitourinary disease, and have diabetes and end-stage kidney disease. Cluster 1 patients had the lowest kidney function, serum bicarbonate, and hemoglobin, but highest serum potassium and prevalence of acute kidney injury. In contrast, cluster 2 patients were the oldest and were more likely to be admitted for respiratory disease, have coronary artery disease, congestive heart failure, stroke, and chronic obstructive pulmonary disease. Cluster 2 patients had lowest serum sodium and serum chloride, but highest serum bicarbonate. Cluster 1 patients had the highest hospital mortality and one-year mortality, followed by cluster 2 and cluster 3, respectively. Conclusion: We identified three clinically distinct phenotypes with differing mortality risks in a heterogeneous cohort of hospitalized hyponatremic patients using an unsupervised machine learning approach.