Winnow CEO Marc Zornes and Iberostar Group's Dr. Megan Morikawa discuss how artificial intelligence can target food waste in commercial kitchens -- and improve both business efficiency and global sustainability. Food waste makes up an estimated 30% to 40% of the food supply, according to the U.S. Department of Agriculture -- and now a London company is using artificial intelligence in an attempt to address the problem. Winnow, a food waste solution company, has developed an AI-powered system that aims to reduce food waste in commercial kitchens worldwide. CEO Marc Zornes said the company's tech can measure the foods that get tossed daily using machine learning and a camera. "We use computer vision to identify what's being wasted in real time, literally as the food's being thrown away," he told Fox News Digital in an interview.

We study a mistake-driven variant of an on-line Bayesian learn(cid:173) ing algorithm (similar to one studied by Cesa-Bianchi, Helmbold, and Panizza [CHP96]). This variant only updates its state (learns) on trials in which it makes a mistake. The algorithm makes binary classifications using a linear-threshold classifier and runs in time lin(cid:173) ear in the number of attributes seen by the learner. We have been able to show, theoretically and in simulations, that this algorithm performs well under assumptions quite different from those embod(cid:173) ied in the prior of the original Bayesian algorithm. It can handle situations that we do not know how to handle in linear time with Bayesian algorithms.

In this paper, we present Committee, a new multi-class learning algo(cid:173) rithm related to the Winnow family of algorithms. Committee is an al(cid:173) gorithm for combining the predictions of a set of sub-experts in the on(cid:173) line mistake-bounded model oflearning. A sub-expert is a special type of attribute that predicts with a distribution over a finite number of classes. Committee learns a linear function of sub-experts and uses this function to make class predictions. We provide bounds for Committee that show it performs well when the target can be represented by a few relevant sub-experts.

We study online learning in Boolean domains using kernels which cap- ture feature expansions equivalent to using conjunctions over basic fea- tures. We demonstrate a tradeoff between the computational efficiency with which these kernels can be computed and the generalization abil- ity of the resulting classifier. We first describe several kernel functions which capture either limited forms of conjunctions or all conjunctions. We show that these kernels can be used to efficiently run the Percep- tron algorithm over an exponential number of conjunctions; however we also prove that using such kernels the Perceptron algorithm can make an exponential number of mistakes even when learning simple func- tions. We also consider an analogous use of kernel functions to run the multiplicative-update Winnow algorithm over an expanded feature space of exponentially many conjunctions.

We establish a mistake bound for an ensemble method for classification based on maximizing the entropy of voting weights subject to margin constraints. The bound is the same as a general bound proved for the Weighted Majority Algorithm, and similar to bounds for other variants of Winnow. We prove a more refined bound that leads to a nearly opti- mal algorithm for learning disjunctions, again, based on the maximum entropy principle. We describe a simplification of the on-line maximum entropy method in which, after each iteration, the margin constraints are replaced with a single linear inequality. The simplified algorithm, which takes a similar form to Winnow, achieves the same mistake bounds.

When it comes to designing user experiences with our systems, the less, the better. We're overwhelmed, to put it mildly, with demands and stimuli. There are millions of apps, applications and websites begging for our attention, and once we have a particular app, application and website up, we still are bombarded by links and choices. Artificial intelligence is offering relief on this front. User experience, driven by AI, may help winnow down a firehose of choices and information needed at the moment down to a gently flowing fountain.

This World Food Day, Emirates Flight Catering (EKFC) has committed to reducing food waste by 35 per cent across its central operations in Dubai. To reach its ambitious goal, EKFC has engaged Winnow to gradually roll out an advanced food waste management system in its state-of-the-art catering facilities. Leveraging artificial intelligence and machine learning, the tool will enable EKFC to automatically monitor and control food waste for the benefit of its customers, its people and broader communities. Saeed Mohammed, chief executive of Emirates Flight Catering, said: "We are dedicated to investing in the latest technologies to optimise our operations and minimise our environmental footprint. "Food waste management has always been an area of focus for EKFC and we have already achieved remarkable results through improved data collection and reporting.

We're overwhelmed, to place it mildly, with calls for and stimuli. There are tens of millions of apps, purposes and web sites begging for our consideration, and as soon as we've a specific app, utility and web site up, we nonetheless are bombarded by hyperlinks and decisions. Every single day, each hour, each minute, it is a firehose. Synthetic intelligence is providing reduction on this entrance. Person expertise, pushed by AI, could assist winnow down a firehose of decisions and knowledge wanted for the time being all the way down to a gently flowing fountain.

Winnow, the U.K. startup that has developed smart kitchen tech to help commercial kitchens reduce food waste, is disclosing $12 million in Series B funding. Backing the round is Ingka Group (a strategic partner to the IKEA franchisee system), Mustard Seed, Circularity Capital, D: Ax and The Ingenious Group. It follows a recent $8 million loan from The European Investment Bank (EIB), meaning that Winnow has added $20 million of capital in the last month. Counting global clients such as IKEA and the Armani Hotel in Dubai, Winnow is on a mission to offer the hospitality industry technology to help cut down on food waste by making commercial kitchens'smarter.' Its latest Winnow Vision product automates waste tracking by using computer vision to track what food is being discarded and therefore enabling kitchens to make better inventory decisions.

Food waste is a $1 trillion problem – costing the world over 1% of global GDP. We're dead set on solving the problem and looking for people to help us achieve our mission. We, at Winnow, believe that food is far too valuable to waste, and that technology can transform the way we produce food. Our team is made of people who all share a passion for food and technology. Winnow was founded in London in 2013 to help the hospitality industry prevent food waste through internet of things tools in the kitchen. We have worked with hundreds of sites and are operating in over 30 countries around the world supported by our offices in London, Dubai, Shanghai, Singapore, Romania and North America.