Since 2010, we have built and maintained LensKit, an open-source toolkit for building, researching, and learning about recommender systems. We have successfully used the software in a wide range of recommender systems experiments, to support education in traditional classroom and online settings, and as the algorithmic backend for user-facing recommendation services in movies and books. This experience, along with community feedback, has surfaced a number of challenges with LensKit's design and environmental choices. In response to these challenges, we are developing a new set of tools that leverage the PyData stack to enable the kinds of research experiments and educational experiences that we have been able to deliver with LensKit, along with new experimental structures that the existing code makes difficult. The result is a set of research tools that should significantly increase research velocity and provide much smoother integration with other software such as Keras while maintaining the same level of reproducibility as a LensKit experiment. In this paper, we reflect on the LensKit project, particularly on our experience using it for offline evaluation experiments, and describe the next-generation LKPY tools for enabling new offline evaluations and experiments with flexible, open-ended designs and well-tested evaluation primitives.

In many online applications interactions between a user and a web-service are organized in a sequential way, e.g., user browsing an e-commerce website. In this setting, recommendation system acts throughout user navigation by showing items. Previous works have addressed this recommendation setup through the task of predicting the next item user will interact with. In particular, Recurrent Neural Networks (RNNs) has been shown to achieve substantial improvements over collaborative filtering baselines. In this paper, we consider interactions triggered by the recommendations of deployed recommender system in addition to browsing behavior. Indeed, it is reported that in online services interactions with recommendations represent up to 30\% of total interactions. Moreover, in practice, recommender system can greatly influence user behavior by promoting specific items. In this paper, we extend the RNN modeling framework by taking into account user interaction with recommended items. We propose and evaluate RNN architectures that consist of the recommendation action module and the state-action fusion module. Using real-world large-scale datasets we demonstrate improved performance on the next item prediction task compared to the baselines.

Conditional preference networks (CP-nets) are a graphical representation of a person's (conditional) preferences over a set of discrete variables. In this paper, we introduce a novel method of quantifying preference for any given outcome based on a CP-net representation of a user's preferences. We demonstrate that these values are useful for reasoning about user preferences. In particular, they allow us to order (any subset of) the possible outcomes in accordance with the user's preferences. Further, these values can be used to improve the efficiency of outcome dominance testing. That is, given a pair of outcomes, we can determine which the user prefers more efficiently. Through experimental results, we show that this method is more effective than existing techniques for improving dominance testing efficiency. We show that the above results also hold for CP-nets that express indifference between variable values.

This paper presents a Discriminative Deep Dyna-Q (D3Q) approach to improving the effectiveness and robustness of Deep Dyna-Q (DDQ), a recently proposed framework that extends the Dyna-Q algorithm to integrate planning for task-completion dialogue policy learning. To obviate DDQ's high dependency on the quality of simulated experiences, we incorporate an RNN-based discriminator in D3Q to differentiate simulated experience from real user experience in order to control the quality of training data. Experiments show that D3Q significantly outperforms DDQ by controlling the quality of simulated experience used for planning. The effectiveness and robustness of D3Q is further demonstrated in a domain extension setting, where the agent's capability of adapting to a changing environment is tested.

Recommendation algorithms are widely adopted in marketplaces to help users find the items they are looking for. The sparsity of the items by user matrix and the cold-start issue in marketplaces pose challenges for the off-the-shelf matrix factorization based recommender systems. To understand user intent and tailor recommendations to their needs, we use deep learning to explore various heterogeneous data available in marketplaces. This paper summarizes five lessons we learned from experimenting with state-of-the-art deep learning recommenders at the leading Norwegian marketplace \textit{FINN.no}. We design a hybrid recommender system that takes the user-generated contents of a marketplace (including text, images and meta attributes) and combines them with user behavior data such as page views and messages to provide recommendations for marketplace items. Among various tactics we experimented with, the following five show the best impact: staged training instead of end-to-end training, leveraging rich user behaviors beyond page views, using user behaviors as noisy labels to train embeddings, using transfer learning to solve the unbalanced data problem, and using attention mechanisms in the hybrid model. This system is currently running with around 20\% click-through-rate in production at \textit{FINN.no} and serves over one million visitors everyday.

Many well-established recommender systems are based on representation learning in Euclidean space. In these models, matching functions such as the Euclidean distance or inner product are typically used for computing similarity scores between user and item embeddings. This paper investigates the notion of learning user and item representations in Hyperbolic space. In this paper, we argue that Hyperbolic space is more suitable for learning user-item embeddings in the recommendation domain. Unlike Euclidean spaces, Hyperbolic spaces are intrinsically equipped to handle hierarchical structure, encouraged by its property of exponentially increasing distances away from origin. We propose HyperBPR (Hyperbolic Bayesian Personalized Ranking), a conceptually simple but highly effective model for the task at hand. Our proposed HyperBPR not only outperforms their Euclidean counterparts, but also achieves state-of-the-art performance on multiple benchmark datasets, demonstrating the effectiveness of personalized recommendation in Hyperbolic space.

Labor Day means a lot of things to a lot of people. For some, it honors America's strong labor force, and for others it means the autumn semester is starting and summer is over. For tech buyers, it's one of the best times to find good deals. A number of retailers are holding labor day sales for TVs, laptops, and other personal electronics. With some help from the team at TechBargains, we've highlighted some of the very best sales below.

Saint Louis University (SLU) has rolled out 2,300 Alexa-powered Echo Dot virtual assistants to all of its student living spaces to provide answers to university-related queries about events, speakers on campus and more. The university also plans to extend use of the artificial intelligence assistant into classrooms and meeting rooms in future and aims to use the technology to support workplace productivity for its faculty staff, according to CIO, David Hakanson. Students arriving at SLU this month can access a custom skill that answers questions relating to university services, such as "When does the library open?" or "Where is the registrar's office?" Already, 130 university-related queries can be answered using the Alexa, and the university is working on more. The idea is that Alexa can reduce the time it takes for students to access information online from seconds to minutes.

At CES 2017, Earin unveiled a revamped version of its true wireless earbuds, called M-2. Over a year and a half later, you can finally get your hands on them. The company was an early entrant into the completely wireless earbud market and although we had reservations about the earlier model, perhaps some of the features M-2 packs in could mean Earin has improved its product from the previous generation. The latest earbuds have Google Assistant integration, which you can activate with a long press. They can also handle calls (using a quartet of microphones that filter out background noise) and let you control your music.

Changing the channel on your television with only a spoken command will soon be possible for the owners of a new range of Toshiba televisions. Starting in 2019, the Japanese firm's OLED, 4K HDR, and Full HD Smart TV products will boast access to Amazon's Alexa voice assistant. Viewers will be able to request channel changes, volume increases and decreases and more by starting their spoken command with'Alexa'. This signals to the always-listening voice assistant that you are about to issue an instruction, which is recorded and parsed in the cloud by Amazon. Toshiba sets will have access to all 45,000 skills already available for Alexa, allowing television viewers to control smart lightbulbs, buy items from Amazon, order takeaways, and check the latest news and weather updates.