Cross-device Federated Learning (FL) faces significant challenges where low-end clients that could potentially make unique contributions are excluded from training large models due to their resource bottlenecks. Recent research efforts have focused on model-heterogeneous FL, by extracting reduced-size models from the global model and applying them to local clients accordingly. Despite the empirical success, general theoretical guarantees of convergence on this method remain an open question. This paper presents a unifying framework for heterogeneous FL algorithms with online model extraction and provides a general convergence analysis for the first time. In particular, we prove that under certain sufficient conditions and for both IID and non-IID data, these algorithms converge to a stationary point of standard FL for general smooth cost functions. Moreover, we introduce the concept of minimum coverage index, together with model reduction noise, which will determine the convergence of heterogeneous federated learning, and therefore we advocate for a holistic approach that considers both factors to enhance the efficiency of heterogeneous federated learning.

Cross-device Federated Learning (FL) faces significant challenges where low-end clients that could potentially make unique contributions are excluded from training large models due to their resource bottlenecks. Recent research efforts have focused on model-heterogeneous FL, by extracting reduced-size models from the global model and applying them to local clients accordingly. Despite the empirical success, general theoretical guarantees of convergence on this method remain an open question. This paper presents a unifying framework for heterogeneous FL algorithms with online model extraction and provides a general convergence analysis for the first time. In particular, we prove that under certain sufficient conditions and for both IID and non-IID data, these algorithms converge to a stationary point of standard FL for general smooth cost functions.

One of the most important initiatives for creating a successful artificial intelligence/machine learning (AI/ML) model is ensuring the data you're using is high quality and clean. That is complete, correct, and relevant to the problem you're trying to solve. Despite the importance of clean data, it can often be overlooked in model creation due to how tedious and time-consuming it can be to review. According to IBM, the lack of clean data, or poor quality data, cost US companies $3.1 trillion in 2016. Accurate models are only built when using clean data.

The COVID-19 crisis forced businesses everywhere to fast track their digital transformation efforts. Faced with the stark choice of becoming a digital-first business, or having no business at all, companies that were previously behind the curve had to implement everything from remote working to entire digital storefronts in a matter of days. According to research by McKinsey, the digital initiatives unleashed in response to the pandemic leapfrogged seven years of progress in a matter of months as companies acted 20 to 25 times faster than they had believed was possible. In the process, this acceleration of digital during the crisis brought about a sea change in executive mindsets with regard to the role of technology in business. Fast forward to today, and corporate leaders are now investing in technology for competitive advantage, refocusing their entire business around cutting-edge technologies, and initiating a business culture where experimentation and innovation is actively encouraged.

The AI industry is playing a dangerous game right now in its embrace of a new generation of citizen developers. On the one hand, AI solution providers, consultants, and others are talking a good talk around "responsible AI." But they're also encouraging a new generation of nontraditional developers to build deep learning, machine learning, natural language processing, and other intelligence into practically everything. A cynic might argue that this attention to responsible uses of technology is the AI industry's attempt to defuse calls for greater regulation. Of course, nobody expects vendors to police how their customers use their products.

In February, after more than a year consulting with a range of experts, the Department of Defense (DoD) released five principles for ethics around artificial intelligence (AI). If AI doesn't meet these standards, the Department has said, it won't be fielded. "The United States, together with our allies and partners, must accelerate the adoption of AI and lead in its national security applications to maintain our strategic position, prevail on future battlefields, and safeguard the rules-based international order," Secretary Mark Esper said in the news release. The principles, which apply to combat and non-combat functions, are that AI must be the following: responsible, equitable, traceable, reliable, and governable. Such guidelines are relatively high level, though, leaving individual departments and agencies on their own to implement what each adjective means for a specific use case.

In February, after more than a year consulting with a range of experts, the Department of Defense (DoD) released five principles for ethics around artificial intelligence (AI). If AI doesn't meet these standards, the Department has said, it won't be fielded. "The United States, together with our allies and partners, must accelerate the adoption of AI and lead in its national security applications to maintain our strategic position, prevail on future battlefields, and safeguard the rules-based international order," Secretary Mark Esper said in the news release. The principles, which apply to combat and non-combat functions, are that AI must be the following: responsible, equitable, traceable, reliable, and governable. Such guidelines are relatively high level, though, leaving individual departments and agencies on their own to implement what each adjective means for a specific use case.

Ensuring that companies make ethical decisions is clearly a boardroom matter. A new study by the IBM Institute for Business Value reported that more than half of the 1,250 executives surveyed believe AI actually can improve their companies' ethical decisions. Yet while 8 out of 10 directors believe the ethical questions raised by the deployment of AI are board-level issues, only 45 percent feel fully prepared to oversee them. AI is different from other technology innovations, affecting how boards approach oversight. Boards are seeking concrete guidance on how to protect the interests of their companies and stakeholders impacted by AI.

As you open the pantry door, you discover that some of the food you purchased yesterday has already gone off or perished – despite the expiry date being weeks away. Feelings of guilt and anger start to wash over you, and rightly so. Food wastage not only affects you economically, but has a widespread ethical and environmental impact. Notwithstanding your own personal loss, there's around 88 million tonnes of food waste in Europe each year, which has associated costs estimated at €143 billion. Put simply by Bernhard Url, executive director of the European Food Safety Authority: "Europe wastes 30% of food, it is an ethical scandal."

Q: Tell us about your work on trusted microelectronics. David Crandall: Our role in this project is to use computer vision and machine learning techniques to help ensure the integrity of the supply chain around microelectronics. One way is to use computer vision to inspect integrated circuits to see whether there is something suspicious that might suggest they are damaged or counterfeit. The goal of computer vision is for computers to be able to understand the visual world the way people do. Computers have been able to take and store pictures for decades, but they haven't been able to know what is in a photo -- what objects and people are in it, what is going on, and what is about to happen.