Clinicians are spending excessive time on EHR systems which leads to burnout and less satisfaction with their work. Physicians spend more than half of their working time on EHR-related tasks1. This averages to be 4.5 hours while in the clinic, and 1.5 hours during their personal time1. The excessive time doctors spend on EHR-related tasks is due to documentation and reporting requirements1. Unfortunately, clinicians end up spending less time seeing patients and communicating with other healthcare professionals because of these burdens. This can lead to medical errors for patients and increased frustration in the workplace.

Grindr is still facing trouble in China. Bloomberg reports the gay dating app has disappeared from Apple's App Store in the country, with researchers at Qimai estimating the software was removed on January 27th. There was no immediate explanation for the departure, but it came just days after China's Cyberspace Administration launched a campaign to purge illegal online material, porn and rumors ahead of the Winter Olympics. We've asked Apple and Grindr for comment. The app's departure came after weeks of glitches with Grindr's service, such as problems adding likes or sending messages.

Gay dating app Grindr has disappeared from multiple app stores in China as authorities tighten control of the country's already heavily policed internet and purge online behaviour the ruling Communist Party dislikes. The country's cyber authority is in the midst of a month-long campaign to root out illegal and sensitive content during the Lunar New Year holiday and February's Winter Olympics. Although the world's most populous nation decriminalised homosexuality in 1997, same-sex marriage is illegal and LGBTQ issues remain taboo. The LGBTQ community is under pressure as censorship of web content combines with a ban on depictions of gay romance in films. Data from mobile research firm Qimai shows that Grindr was removed from Apple's App Store in China on Thursday.

NLP is leveraged to improve search results by filtering and contextualizing them, making the results more relevant for shoppers. This can also be done by focusing on visual elements in a search. Thanks to machine learning, AI software tags, arranges and searches visually for content by categorizing image and/or video features. This technology also allows shoppers to not just find products that match, but those that complement them too. So to purchase an item, a customer need not search and browse endlessly, simply upload an image. It often happens that sales teams fail to follow up with marketing leads, allowing qualified potential customers to get lost in the system.

A 2019 study conducted by DataChildFutures found that 46% of participating Italian households had AI-powered speakers, while 40% of toys were connected to the internet. More recent research suggests that by 2023 more than 275 million intelligent voice assistants, such as Amazon Echo or Google Home, will be installed in homes worldwide. As younger generations grow up interacting with AI-enabled devices, more consideration should be given to the impact of this technology on children, their rights and wellbeing. AI-powered learning tools and approaches are often regarded as critical drivers of innovation in the education sector. Often recognized for its ability to improve the quality of learning and teaching, artificial intelligence is being used to monitor students' level of knowledge and learning habits, such as rereading and task prioritization, and ultimately to provide a personalized approach to learning. Knewton is one example of AI-enabled learning software that identifies knowledge gaps and curates education content in line with user needs.

Maximizing a monotone submodular function is a fundamental task in machine learning. In this paper, we study the deletion robust version of the problem under the classic matroids constraint. Here the goal is to extract a small size summary of the dataset that contains a high value independent set even after an adversary deleted some elements. We present constant-factor approximation algorithms, whose space complexity depends on the rank $k$ of the matroid and the number $d$ of deleted elements. In the centralized setting we present a $(3.582+O(\varepsilon))$-approximation algorithm with summary size $O(k + \frac{d \log k}{\varepsilon^2})$. In the streaming setting we provide a $(5.582+O(\varepsilon))$-approximation algorithm with summary size and memory $O(k + \frac{d \log k}{\varepsilon^2})$. We complement our theoretical results with an in-depth experimental analysis showing the effectiveness of our algorithms on real-world datasets.

I recently presented the plenary session at a pharma conference covering how Artificial Intelligence (AI) is transforming pharma sales and marketing, I provided examples Eularis had completed for pharma client projects. Several of the attendees sent me emails afterwards wanting to know more about the specific examples I gave, which were as varied as our client needs. It was interesting to learn how few of these types of applications they were familiar with, and I thought the readers of my white papers would want to know about them, too. I've written about many of these topics before, and I'm including those links at the end of each section in case you are interested in digging deeper into a specific topic. According to Takeda Pharmaceuticals, the average time taken to diagnose a rare disease without technology is 7.6 years and comes after countless tests and physician visits. This creates a high cost to the healthcare system, not to mention much suffering for the patient. And some cases are even worse.

Ranking, recommendation, and retrieval systems are widely used in online platforms and other societal systems, including e-commerce, media-streaming, admissions, gig platforms, and hiring. In the recent past, a large "fair ranking" research literature has been developed around making these systems fair to the individuals, providers, or content that are being ranked. Most of this literature defines fairness for a single instance of retrieval, or as a simple additive notion for multiple instances of retrievals over time. This work provides a critical overview of this literature, detailing the often context-specific concerns that such an approach misses: the gap between high ranking placements and true provider utility, spillovers and compounding effects over time, induced strategic incentives, and the effect of statistical uncertainty. We then provide a path forward for a more holistic and impact-oriented fair ranking research agenda, including methodological lessons from other fields and the role of the broader stakeholder community in overcoming data bottlenecks and designing effective regulatory environments.

Motivated by online recommendation systems, we propose the problem of finding the optimal policy in multitask contextual bandits when a small fraction $\alpha < 1/2$ of tasks (users) are arbitrary and adversarial. The remaining fraction of good users share the same instance of contextual bandits with $S$ contexts and $A$ actions (items). Naturally, whether a user is good or adversarial is not known in advance. The goal is to robustly learn the policy that maximizes rewards for good users with as few user interactions as possible. Without adversarial users, established results in collaborative filtering show that $O(1/\epsilon^2)$ per-user interactions suffice to learn a good policy, precisely because information can be shared across users. This parallelization gain is fundamentally altered by the presence of adversarial users: unless there are super-polynomial number of users, we show a lower bound of $\tilde{\Omega}(\min(S,A) \cdot \alpha^2 / \epsilon^2)$ {\it per-user} interactions to learn an $\epsilon$-optimal policy for the good users. We then show we can achieve an $\tilde{O}(\min(S,A)\cdot \alpha/\epsilon^2)$ upper-bound, by employing efficient robust mean estimators for both uni-variate and high-dimensional random variables. We also show that this can be improved depending on the distributions of contexts.

Amid rapidly changing technology, many people still associate artificial intelligence (AI) with science-fiction dystopias. But in reality, AI has become an integral part of our daily lives. We now rely on search-engine algorithms and digital assistants like Alexa and Siri for almost everything, including ordering a taxi or finding out how many calories there are in a 10-inch Margherita pizza. The potential of this technology goes beyond its household use. AI is making great strides in the business world, and creativity may be its ultimate moonshot.