Collaborative filtering (CF), as a fundamental approach for recommender systems, is usually built on the latent factor model with learnable parameters to predict users' preferences towards items. However, designing a proper CF model for a given data is not easy, since the properties of datasets are highly diverse. In this paper, motivated by the recent advances in automated machine learning (AutoML), we propose to design a data-specific CF model by AutoML techniques. The key here is a new framework that unifies state-of-the-art (SOTA) CF methods and splits them into disjoint stages of input encoding, embedding function, interaction function, and prediction function. We further develop an easy-to-use, robust, and efficient search strategy, which utilizes random search and a performance predictor for efficient searching within the above framework. In this way, we can combinatorially generalize data-specific CF models, which have not been visited in the literature, from SOTA ones. Extensive experiments on five real-world datasets demonstrate that our method can consistently outperform SOTA ones for various CF tasks. Further experiments verify the rationality of the proposed framework and the efficiency of the search strategy. The searched CF models can also provide insights for exploring more effective methods in the future

Abstract: Conversational agents are a recent trend in human-computer interaction, deployed in multidisciplinary applications to assist the users. In this paper, we introduce "Atreya", an interactive bot for chemistry enthusiasts, researchers, and students to study the ChEMBL database. Atreya is hosted by Telegram, a popular cloud-based instant messaging application. This user-friendly bot queries the ChEMBL database, retrieves the drug details for a particular disease, targets associated with that drug, etc. This paper explores the potential of using a conversational agent to assist chemistry students and chemical scientist in complex information seeking process.

We consider the problem of minimizing regret in an $N$ agent heterogeneous stochastic linear bandits framework, where the agents (users) are similar but not all identical. We model user heterogeneity using two popularly used ideas in practice; (i) A clustering framework where users are partitioned into groups with users in the same group being identical to each other, but different across groups, and (ii) a personalization framework where no two users are necessarily identical, but a user's parameters are close to that of the population average. In the clustered users' setup, we propose a novel algorithm, based on successive refinement of cluster identities and regret minimization. We show that, for any agent, the regret scales as $\mathcal{O}(\sqrt{T/N})$, if the agent is in a `well separated' cluster, or scales as $\mathcal{O}(T^{\frac{1}{2} + \varepsilon}/(N)^{\frac{1}{2} -\varepsilon})$ if its cluster is not well separated, where $\varepsilon$ is positive and arbitrarily close to $0$. Our algorithm is adaptive to the cluster separation, and is parameter free -- it does not need to know the number of clusters, separation and cluster size, yet the regret guarantee adapts to the inherent complexity. In the personalization framework, we introduce a natural algorithm where, the personal bandit instances are initialized with the estimates of the global average model. We show that, an agent $i$ whose parameter deviates from the population average by $\epsilon_i$, attains a regret scaling of $\widetilde{O}(\epsilon_i\sqrt{T})$. This demonstrates that if the user representations are close (small $\epsilon_i)$, the resulting regret is low, and vice-versa. The results are empirically validated and we observe superior performance of our adaptive algorithms over non-adaptive baselines.

Machine Learning, What is machine learning?, How machine learning works, Machine learning methods, Deep learning & more You may also have heard machine learning and AI used interchangeably. AI includes machine learning, but machine learning doesn't fully define AI. Machine learning and AI both have strong engineering components. You find AI and machine learning used in a great many applications today. Artificial Intelligence (AI) is a huge topic today, and it's getting bigger all the time thanks to the success of technologies such as Siri.

Global access to art, culture, and entertainment products – music, movies, books, and more – has undergone fundamental changes over the past 20 years in light of groundbreaking developments in artificial intelligence. For example, users of streaming services like Netflix and Spotify have data collected and analyzed by algorithms to determine their streaming habits – resulting in recommendations that cater to their tastes. But this is only one of the many ways in which AI tools are transforming the arts and culture industries. AI is also being used in the production of music and other art, with algorithms generating photos or writing songs on their own. Warner Music even "signed" an algorithm to a record deal in 2019.

AI is a large part of our world already, affecting online search results and the way we shop. Interest in AI has attracted long-term investments in AI use across several industries, particularly in customer service, medical diagnostics and self-driving vehicles. The increased data available through research has created better algorithms which have enabled more complex AI systems that improve a user's experience with search engines and online translation tools, but also means that businesses can make far more focused sales and marketing drives to customers and financial markets have virtual assistants able to deal with more than the simplest of requests. AI system improvements will involve the processing of massive amounts of data which needs improved computing power and better algorithms and tools. Using cryptography and blockchain has made it easier to build these advances since they can publicly share data whilst keeping company information confidential.

Designing an effective loss function plays a crucial role in training deep recommender systems. Most existing works often leverage a predefined and fixed loss function that could lead to suboptimal recommendation quality and training efficiency. Some recent efforts rely on exhaustively or manually searched weights to fuse a group of candidate loss functions, which is exceptionally costly in computation and time. They also neglect the various convergence behaviors of different data examples. In this work, we propose an AutoLoss framework that can automatically and adaptively search for the appropriate loss function from a set of candidates. To be specific, we develop a novel controller network, which can dynamically adjust the loss probabilities in a differentiable manner. Unlike existing algorithms, the proposed controller can adaptively generate the loss probabilities for different data examples according to their varied convergence behaviors. Such design improves the model's generalizability and transferability between deep recommender systems and datasets. We evaluate the proposed framework on two benchmark datasets. The results show that AutoLoss outperforms representative baselines. Further experiments have been conducted to deepen our understandings of AutoLoss, including its transferability, components and training efficiency.

Fox News Flash top headlines are here. Check out what's clicking on Foxnews.com. A Nebraska man who murdered and dismembered a young woman he met on Tinder and then slashed his own neck in court during his trial was sentenced to death this week. Aubrey Trail, a 54-year-old thief and con man, was convicted of strangling 24-year-old Sydney Loofe with an electrical cord in 2017, then cutting her body into 14 pieces that he dumped in various rural roadside ditches. At trial, witnesses testified that Trail and his girlfriend, Bailey Boswell, 27, had solicited them for group sex and talked of the occult and gaining "powers" through killing.

Years ago, we searched the web, bought new gadgets, and typed in our email addresses without much thought. As far as accounts went, "Hey if it's free, sign me up," we thought. Fast forward to now, and you can't go online or turn on the news without hearing about the control Big Tech has on our lives – and the growing resentment around it. Probably due to government initiatives, tech companies are making changes to address these concerns. You can now password protect the page that reveals all your Google searches and other activity.

Researchers at the University of Michigan have proposed a way to have your privacy cake and eat your home automation too. They've found a means of using a voice-activated smart speaker system without it having to listen to everything you say – and no, it's not "pressing a button." "There are a lot of situations where we want our home automation system or our smart speaker to understand what's going on in our home, but we don't necessarily want it listening to our conversations," said the aptly named Alanson Sample, associate professor of electrical engineering and computer science at the University of Michigan. "And what we've found is that you can have a system that understands what's going on and a hard guarantee that it will never record any audible information." A video showcasing the technology explains.