Federated Learning (FL) is a collaborative learning performed by a moderator that protects data privacy. Existing cross-silo FL solutions seldom address the absence of participating clients during training which can seriously degrade model performances, particularly for unbalanced and non-IID client data. We address this issue by generating secure data digests from the raw data and using them to guide model training at the FL moderator. The proposed FL with data digest (FedDig) framework can tolerate unexpected client absence while preserving data privacy. This is achieved by de-identifying digests by mixing and perturbing the encoded features of the raw data in the feature space. The feature perturbing is performed following the Laplace mechanism of Differential Privacy. We evaluate FedDig on EMNIST, CIFAR-10, and CIFAR-100 datasets. The results consistently outperform three baseline algorithms (FedAvg, FedProx, and FedNova) by large margins in multiple client absence scenarios.

A new product to make ML easier and how marketing and financial services benefit from ML. Beekeepers and nonprofits (among others) are trying out a new product from Microsoft that aims to help nontechnical users get started with image classification ML. Automation and machine learning are helping marketers scale up their efforts. A new study highlights how AI and machine learning are being used across financial services.

Using AI to manage COVID-19 risks and applying predictive models for multiple kinds of retail. U.K. retailers are applying AI to track customer feedback and manage new risks caused by the COVID-19 pandemic. Nike is using predictive models to optimize warehouse inventory. A U.K. retail group is increasing its investment in AI and predictive analytics after a trial run reports great results.

How machine learning and AI are being used to cut emissions, picture the past, and study DNA. This company claims their AI platform can cut carbon dioxide emissions by improving buildings' efficiency. Read how an artist used machine learning to extrapolate realistic portraits of ancient Roman emperors. Researchers at the University of California San Diego have used machine learning to solve a long-standing question about gene activation in humans.

Read about new big data practices, how big data is being applied for good purposes, and how AI and NLP are changing the use of big data. As the use of big data continues to grow, many new practices are emerging for 2019. Read about five data science projects leveraging big data to tackle huge projects for social good. Applying AI and natural language processing is driving new value and innovation from big data.

How AI, machine learning, and big data analytics can help cybersecurity, and examples of real applications for AI and machine learning. A new survey indicates that many cybersecurity professionals believe that AI and machine learning techniques provide major benefits. Big data analytics can be an important tool for defeating cyberattacks. Here are fifteen examples of real-world applications for AI and ML that are improving enterprises today.

How to use data science for hiring and retention, why you should avoid relying on one employee's knowledge, and what the next big data role to hire might be. Careful use of data science can improve your hiring methods and help you keep good employees around. Many enterprises have that one employee who knows everything about the data and seems to work magic, but it's dangerous to rely on that kind of personal knowledge over the long term. Data engineers are in high demand because they build the infrastructure needed to perform analysis and create machine learning models.

Detecting future fraud with machine learning, what cybersecurity has in common with data science, and why all security should be proactive. Researchers in cybersecurity are working on predictive machine learning algorithms that may be able to identify future fraud victims. Most cybersecurity professionals are doing data science, although they often don't call it that. A good cybersecurity posture should always be proactive.

The definitions of AI and machine learning, how to tell whether you need ML, and what the spread of Internet bots could mean. If you're hoping to use them in your business, it's important to know the difference between machine learning (ML) and artificial intelligence (AI). Ask these questions before you decide to invest resources. Bots are everywhere in today's online world. What happens if this trend continues?

AI and machine learning are susceptible to flawed data and unseen bias everywhere, but start-ups in emerging markets should be especially careful. Check this list of best practices to follow. If you're not seeing the outcomes you want from machine learning, you may have problems in the data sets used to train the algorithms. This blog post explains why so many teams fail to notice such problems and how to fix them. Researchers with IBM created a system that could improve machine learning models by teaching them to detect what is missing from a data set.