In this paper, we propose a robust sequential learning strategy for training large-scale Recommender Systems (RS) over implicit feedback mainly in the form of clicks. Our approach relies on the minimization of a pairwise ranking loss over blocks of consecutive items constituted by a sequence of non-clicked items followed by a clicked one for each user. Parameter updates are discarded if for a given user the number of sequential blocks is below or above some given thresholds estimated over the distribution of the number of blocks in the training set. This is to prevent from an abnormal number of clicks over some targeted items, mainly due to bots; or very few user interactions. Both scenarios affect the decision of RS and imply a shift over the distribution of items that are shown to the users. We provide a theoretical analysis showing that in the case where the ranking loss is convex, the deviation between the loss with respect to the sequence of weights found by the proposed algorithm and its minimum is bounded. Furthermore, experimental results on five large-scale collections demonstrate the efficiency of the proposed algorithm with respect to the state-of-the-art approaches, both regarding different ranking measures and computation time.

Whether through chatbots or voice assistants, bank customers are beginning to experience Banking AI and are responding favourably. Initially, the focus of these interfaces is similar to the task-oriented nature of channels that banks already provide. Though conversational interfaces have yet to attain the level of adoption expected in the banking space and has not yet matured as mobile has, there are already a variety of approaches for establishing a conversational banking capability. Going forward, I believe there will be at least four distinct banking models supported by apps and conversational interfaces. The majority of banks will begin their journey in the conversational banking space by implementing a conversational interface as an additional channel with a task-based focus.

Facebook wants to create its own chips to deliver the type of computing speeds necessary to take the next leap forward in artificial intelligence, according to a report in the Financial Times. The company also wants to make a digital assistant – similar to Apple's Siri or Amazon's Alexa – that can have conversations with users and that has "common sense." Yann LeCun, the company's chief AI scientist and a pivotal figure in modern AI, said the company also wants to use AI to help monitor video in real time and assist human moderators with content selection. Though the company is working with chip companies like Intel on the new chips, it is not ruling out the possibility of developing its own. "Facebook has been known to build its hardware when required -- build its own ASIC, for instance. If there's any stone unturned, we're going to work on it," LeCun said.

Collaborative filtering (CF)[1] based on the alternating least squares (ALS) technique[2] is another algorithm used to generate recommendations. It produces automatic predictions (filtering) about the interests of a user by collecting preferences from many other users (collaborating). The underlying assumption of the collaborative filtering approach is that if a person A has the same opinion as a person B on an issue, A is more likely to have B's opinion on a different issue than a randomly chosen person. This algorithm gained a lot of traction in the data science community after it was used by the team winner of the Netflix Prize. The CF algorithm has also been implemented in Spark MLlib[3] with the aim to address fast execution on very large datasets.

The practice of collecting basic demographic information from customers to create a successful business marketing strategy is one of the past. In recent times, there has been a major shift in the way that businesses interact with their customers. The digital space has spread so far and wide that it has had a lasting influence on virtually everything we do. As a result, the conventional approaches to marketing that were prevalent even as early as a few years ago are considered severely ineffective today. The rapidly growing popularity of Big Data means that marketers need to embrace sophisticated approaches to processes and perform in-depth analysis of customer data, preferably in real-time.

Chatbots and AI assistants need further development to reach the point where they can fully understand the nuances of language and engage in real conversations with people. But good design can help organizations overcome the limitations of existing conversational AI bots, according to developers at companies on the leading edge of the trend. "I think it's a myth that we'll have to wait until the best AI materializes to improve assistants," said Cathy Pearl, head of conversation design outreach at Google. "Actually, we can do a lot now with good design principles and a little technology that we have now." Conversational AI was one of the big discussion topics at the 2019 Re•Work Deep Learning Summit in San Francisco.

Despite the overall tone, which was a tad too optimistic for my taste, I found the video insightful and recommend a watch. OpenAI's new algorithm learns from mistakes

Large-scale industrial recommender systems are usually confronted with computational problems due to the enormous corpus size. To retrieve and recommend the most relevant items to users under response time limits, resorting to an efficient index structure is an effective and practical solution. Tree-based Deep Model (TDM) for recommendation \cite{zhu2018learning} greatly improves recommendation accuracy using tree index. By indexing items in a tree hierarchy and training a user-node preference prediction model satisfying a max-heap like property in the tree, TDM provides logarithmic computational complexity w.r.t. the corpus size, enabling the use of arbitrary advanced models in candidate retrieval and recommendation. In tree-based recommendation methods, the quality of both the tree index and the trained user preference prediction model determines the recommendation accuracy for the most part. We argue that the learning of tree index and user preference model has interdependence. Our purpose, in this paper, is to develop a method to jointly learn the index structure and user preference prediction model. In our proposed joint optimization framework, the learning of index and user preference prediction model are carried out under a unified performance measure. Besides, we come up with a novel hierarchical user preference representation utilizing the tree index hierarchy. Experimental evaluations with two large-scale real-world datasets show that the proposed method improves recommendation accuracy significantly. Online A/B test results at Taobao display advertising also demonstrate the effectiveness of the proposed method in production environments.

Amazon, Apple and Google voice assistants and smart TVs could find out if a person is cheating on their partner, a data expert has claimed. Smart gadgets, which are used by one in ten people around the UK, can harvest enough data to work out the dynamics of a relationship, they say. They have the potential to record saucy conversations and analyse location data to discover secret affairs. The virtual assistants can show when occupants are in a building, or for example, share a bedroom, by using sensor logs and smart meters. There is already ambiguity when it comes to these companies privacy policies and how they collect and use people's data.

A few weeks after being pulled because of connectivity issues, Amazon's Echo Wall Clock is once again available. The clock can once again be purchased for $29.99 from Amazon. A spokesperson for Amazon confirmed the availability of the clock to Engadget. The company also said that a software update will be made available for customers who have already purchased the clock. That update will be received automatically when the clock is connected to an Echo device.