Recommender system is an important component of many web services to help users locate items that match their interests. Several studies showed that recommender systems are vulnerable to poisoning attacks, in which an attacker injects fake data to a given system such that the system makes recommendations as the attacker desires. However, these poisoning attacks are either agnostic to recommendation algorithms or optimized to recommender systems that are not graph-based. Like association-rule-based and matrix-factorization-based recommender systems, graph-based recommender system is also deployed in practice, e.g., eBay, Huawei App Store. However, how to design optimized poisoning attacks for graph-based recommender systems is still an open problem. In this work, we perform a systematic study on poisoning attacks to graph-based recommender systems. Due to limited resources and to avoid detection, we assume the number of fake users that can be injected into the system is bounded. The key challenge is how to assign rating scores to the fake users such that the target item is recommended to as many normal users as possible. To address the challenge, we formulate the poisoning attacks as an optimization problem, solving which determines the rating scores for the fake users. We also propose techniques to solve the optimization problem. We evaluate our attacks and compare them with existing attacks under white-box (recommendation algorithm and its parameters are known), gray-box (recommendation algorithm is known but its parameters are unknown), and black-box (recommendation algorithm is unknown) settings using two real-world datasets. Our results show that our attack is effective and outperforms existing attacks for graph-based recommender systems. For instance, when 1% fake users are injected, our attack can make a target item recommended to 580 times more normal users in certain scenarios.

Would you let an Artificial Intelligence make decisions on for behalf? With all the talk about AI and calls for its regulation (apparently AI will kill us all someday,) I've been thinking about under what circumstances we might take away the decision making from the machines (call it AI). But still we use them to our advantage (as much as we can) -- that's not as crazy as it sounds, because humanity has been doing that for years already. It's everywhere, and there is no hiding from the fact that it is here to stay. From the incredibly-friendly voice of Apple's personal assistant, Siri, to movies like Ex-Machina, Al has always excited me more than anything else.

"AI is one of the most important things humanity is working on. It is more profound than electricity or fire…We have learned to harness fire for the benefits of humanity but we had to overcome its downsides too.….AI is really important, but we have to be concerned about it." While science fiction often portrays AI as robots with human-like characteristics, AI can encompass anything from Google's search algorithms to IBM's Watson, to autonomous weapons. Everyone is excited about AI, and everyone has a view on AI. AI is no longer the preserve of an Alex Garland screenplay, it's a reality and being used in multiple ways.

Emerging technology trends clearly point to a future encompassing screen-less interactions between businesses and consumers, with voice, augmented and virtual reality, wearable devices, and artificial intelligence, gradually but definitely removing the traditional graphic user interface (GUI) from the equation. The next decade is expected to be even more disruptive based on the methodologies used by customers to interact with brands. A closer glimpse of the consumer landscape, reveals irrefutable enthusiasm for artificial intelligence (AI) as compared to other upcoming technologies. However, the technology is still in the experimental phase. Even though the majority of enterprise leaders consider AI to be a business advantage, many organizations are taciturn to trust AI to the extent of deferring implementation and hence are yet to benefit from the technology's promising capabilities.

In the documentary Swiped, filmmaker Nancy Jo Sales investigates how dating apps have created unintended consequences in actual relationships. In the documentary Swiped, filmmaker Nancy Jo Sales investigates how dating apps have created unintended consequences in actual relationships. For some of the 40 million or so Americans who currently use online dating apps like Tinder, Bumble and Hinge, the findings of the new HBO documentary Swiped might be intuitively obvious. But for others, there may still be revelations aplenty in the film, which is subtitled Hooking Up in the Digital Age. It's about how these apps may change how we think about relationships -- and it doesn't paint a positive picture.

Comedian Norm MacDonald just co-created a new video dating app, LOKO, but he's not ready to put himself out there on it. "I am not using it," he told USA TODAY this week. MacDonald is best known for his years on Saturday Night Live and sitcoms like The Norm Show. He's got a new gig for Netflix, a talk show called Norm MacDonald Has a Show, debuting on Sept. 14th. The LOKO app came out of a corporate gig, where he met Canadian entrepreneur Vivek Jain, who complained about his lack of success finding dates.

What features are you hoping to see in the new iPhone? USA TODAY's Jefferson Graham provides his wishlist for Tim Cook and Apple. We've got the kickoff to the gadget-selling season Wednesday with the introduction of new iPhones and possibly more Apple products. I love to geek out as much as the next guy, so thanks for inviting me. You've got quite a sales job ahead of you, though.

Jefferson Graham offers tips on how to listen to the Talking Tech podcast via apps, speakers, TVs and the car. The Talking Tech podcast is available for you every day with a quick hit on the latest tech news, gadget reviews, opinion on tech trends and interviews with insiders. On this page, you'll find quick links to all of our shows. In September so far, we've covered everything from how to watch NFL football games on streaming, new tools added to Amazon's Echo to offer voice-activated concert and touring information and our tips on why you shouldn't buy a new or older iPhone until after September 21st. We had comedian Norm MacDonald visit Talking Tech to tell about the new dating app he created with a friend.

While Google has yet to launch its digital wellbeing features on Android phones, it's already thinking of how to bring them to the smart speaker in your living room. The 9to5Google team has discovered material in the the latest Google app for Android that points to these digital health features coming to Home speakers and other Assistant-equipped devices. While the exact functionality isn't clear, there's a Downtime feature that could stop people from using Home at certain times of the day, such as when they're winding down at the end of the night. That could be particularly helpful if you have a bad habit of playing music or asking questions when you should be getting to bed. "Filters," meanwhile, appears set to replace Family Mode and may limit what kids are allowed to search for, or prevent them from playing some voice-based games.

Although modern recommendation systems can exploit the structure in users' item feedback, most are powerless in the face of new users who provide no structure for them to exploit. In this paper we introduce ImplicitCE, an algorithm for recommending items to new users during their sign-up flow. ImplicitCE works by transforming users' implicit feedback towards auxiliary domain items into an embedding in the target domain item embedding space. ImplicitCE learns these embedding spaces and transformation function in an end-to-end fashion and can co-embed users and items with any differentiable similarity function. To train ImplicitCE we explore methods for maximizing the correlations between model predictions and users' affinities and introduce Sample Correlation Update, a novel and extremely simple training strategy. Finally, we show that ImplicitCE trained with Sample Correlation Update outperforms a variety of state of the art algorithms and loss functions on both a large scale Twitter dataset and the DBLP dataset.