Biscotti once fed Roman navies and Christopher Columbus's expeditions Long before it met espresso, this crunchy pastry kept sailors fed. Roman writer Pliny the Elder was the first writer to mention biscotti in 77 CE. Breakthroughs, discoveries, and DIY tips sent every weekday. Step into a typical Italian restaurant in the U.S. and you'll likely find "biscotti" on the menu. Typically served with a glass of sweet wine or cappuccino, these log-shaped crunchy cookies are a beloved treat that most of us associate with cozy dinners and Little Italy.

Federated learning (FL) is an emerging promising paradigm of privacy-preserving machine learning (ML). An important type of FL is cross-silo FL, which enables a small scale of organizations to cooperatively train a shared model by keeping confidential data locally and aggregating weights on a central parameter server. However, the central server may be vulnerable to malicious attacks or software failures in practice. To address this issue, in this paper, we propose DeFL, a novel decentralized weight aggregation framework for cross-silo FL. DeFL eliminates the central server by aggregating weights on each participating node and weights of only the current training round are maintained and synchronized among all nodes. We use Multi-Krum to enable aggregating correct weights from honest nodes and use HotStuff to ensure the consistency of the training round number and weights among all nodes. Besides, we theoretically analyze the Byzantine fault tolerance, convergence, and complexity of DeFL. We conduct extensive experiments over two widely-adopted public datasets, i.e. CIFAR-10 and Sentiment140, to evaluate the performance of DeFL. Results show that DeFL defends against common threat models with minimal accuracy loss, and achieves up to 100x reduction in storage overhead and up to 12x reduction in network overhead, compared to state-of-the-art decentralized FL approaches.

We are excited to bring Transform 2022 back in-person July 19 and virtually July 20 - 28. Join AI and data leaders for insightful talks and exciting networking opportunities. Many are familiar with the idea of factory automation, but what about'hyperautomation'? And, how about the rise of autonomous factories, with systems that make their own decisions about things like quality control and line speed? Both concepts, driven by artificial intelligence (AI) technologies, are coming soon to manufacturers, and are being closely tracked by many industry watchers. They're also both expected to revolutionize how factories function.

Revenue from AI software soared more than 63 per cent to $846m (£664m) last year amid a surge in firms investing in the new technology. Robotic process automation (RPA), which provides AI tools for businesses, is now the fastest-growing segment of the enterprise software market, with revenue set to reach $1.3bn in 2019, according to research firm Gartner. "The RPA market has grown since our last forecast, driven by digital business demands as organisations look for'straight-through' processing," said Fabrizio Biscotti, research vice president at Gartner. "Competition is intense, with nine of the top 10 vendors changing market share position in 2018." The top five RPA vendors controlled 47 per cent of the market in 2018, though the vendors ranked sixth and seventh posted trip-digit revenue growth in a sign of heightening competition.

Federated Learning is the current state of the art in supporting secure multi-party ML: data is maintained on the owner's device and is aggregated through a secure protocol. However, this process assumes a trusted centralized infrastructure for coordination and clients must trust that the central service does not maliciously omit client contributions or use the byproducts of client data. As a response, we propose Biscotti: a fully decentralized P2P approach to multi-party ML, which uses blockchain and crypto primitives to coordinate a privacy-preserving ML process between peering clients. Our evaluation demonstrates that Biscotti is scalable, fault tolerant, and defends against known attacks. For example, Biscotti is able to protect the performance of the global model at scale even when 45% of adversaries are trying to poison the model. The implementation can be found at: https://github.com/DistributedML/Biscotti