Recently, self-learning methods based on user satisfaction metrics and contextual bandits have shown promising results to enable consistent improvements in conversational AI systems. However, directly targeting such metrics by off-policy bandit learning objectives often increases the risk of making abrupt policy changes that break the current user experience. In this study, we introduce a scalable framework for supporting fine-grained exploration targets for individual domains via user-defined constraints. For example, we may want to ensure fewer policy deviations in business-critical domains such as shopping, while allocating more exploration budget to domains such as music. Furthermore, we present a novel meta-gradient learning approach that is scalable and practical to address this problem. The proposed method adjusts constraint violation penalty terms adaptively through a meta objective that encourages balanced constraint satisfaction across domains. We conduct extensive experiments using data from a real-world conversational AI on a set of realistic constraint benchmarks. Based on the experimental results, we demonstrate that the proposed approach is capable of achieving the best balance between the policy value and constraint satisfaction rate.

In this paper, we research the new topic of object effects recommendation in micro-video platforms, which is a challenging but important task for many practical applications such as advertisement insertion. To avoid the problem of introducing background bias caused by directly learning video content from image frames, we propose to utilize the meaningful body language hidden in 3D human pose for recommendation. To this end, in this work, a novel human pose driven object effects recommendation network termed PoseRec is introduced. PoseRec leverages the advantages of 3D human pose detection and learns information from multi-frame 3D human pose for video-item registration, resulting in high quality object effects recommendation performance. Moreover, to solve the inherent ambiguity and sparsity issues that exist in object effects recommendation, we further propose a novel item-aware implicit prototype learning module and a novel pose-aware transductive hard-negative mining module to better learn pose-item relationships. What's more, to benchmark methods for the new research topic, we build a new dataset for object effects recommendation named Pose-OBE. Extensive experiments on Pose-OBE demonstrate that our method can achieve superior performance than strong baselines.

In this work, we present our journey to revolutionize the personalized recommendation engine through end-to-end learning from raw user actions. We encode user's long-term interest in Pinner- Former, a user embedding optimized for long-term future actions via a new dense all-action loss, and capture user's short-term intention by directly learning from the real-time action sequences. We conducted both offline and online experiments to validate the performance of the new model architecture, and also address the challenge of serving such a complex model using mixed CPU/GPU setup in production. The proposed system has been deployed in production at Pinterest and has delivered significant online gains across organic and Ads applications.

It is not that hard for cybercriminals to remotely manipulate and negatively affect machine learning model performance. Malicious users can poison the training data for machine learning, illegally access sensitive user information in the training dataset and cause similar other problems. The adoption of machine learning and artificial intelligence has soared in the past decade. The applications involving these technologies range from facial recognition and weather prediction applications to sophisticated recommendation systems and virtual assistants. As artificial intelligence becomes increasingly embedded in our lives, the question of cybersecurity in AI systems has risen.

Yesterday, California-based AI firm Adept announced Action Transformer (ACT-1), an AI model that can perform actions in software like a human assistant when given high-level written or verbal commands. It can reportedly operate web apps and perform intelligent searches on websites while clicking, scrolling, and typing in the right fields as if it were a person using the computer. In a demo video tweeted by Adept, the company shows someone typing, "Find me a house in Houston that works for a family of 4. My budget is 600K" into a text entry box. Upon submitting the task, ACT-1 automatically browses Redfin.com in a web browser, clicking the proper regions of the website, typing a search entry, and changing the search parameters until a matching house appears on the screen. It's called Action Transformer (ACT-1) and we taught it to use a bunch of software tools.

Conventionally, salespeople are responsible for juggling tasks like following a playbook, capturing responses, building rapport and updating a customer relationship management (CRM) system during sales calls. As these tend to be repetitive and time-consuming, tedium can quickly set in. The average salesperson spends more than five hours a week updating CRM records, according to a Dooly survey. In search of a solution, sales tech entrepreneur Eldad Postan-Koren and cybersecurity practitioner Bar Haleva co-created Winn.AI, an AI-powered assistant designed to help sales teams automatically track, capture and update CRM entries. Winn.AI monitors sales calls and records key data, in theory reducing the need for salespeople to note-take themselves.

Winn.AI handles the busywork of sales, so salespeople can focus on selling. Capture meeting notes & customer data in real-time and update your CRM in a click.

Years ago, artificial intelligence (AI) was only exclusive to scientific research. Nowadays, you can also find AI in the eCommerce industry, as it has become an invaluable tool to many companies in the online space. Numerous businesses use AI to reduce operating expenses, improve analytical insights, and stay ahead of competitors. There's no doubt that AI provides many new opportunities for eCommerce, which is why more and more companies adopt this technology. The growth of AI is at a rapid pace.

Thank you for calling …" A sad sort of human-to-computer stalemate has played out over countless fruitless interactions. Companies adopted IVR (interactive voice response) systems over the last decades, possibly in an attempt to reduce the cost of hiring and training costs for human customer service reps that have high turnover rates. Or, perhaps some forward-thinking executives thought a robot-voiced CSR would make a company appear more'advanced' in comparison to its competitors. Whatever the reason, our earliest conversations with IVR menus and chatbots left most of us humans feeling let down, like we weren't having a conversation at all. Despite the fact that voice recognition and computer speech have improved dramatically in speed and sophistication, it's hard for some of us to shake that feeling that nobody is on the other end of the line to help.

Recommender systems, which offer personalized suggestions to users, power many of today's social media, e-commerce and entertainment. However, these systems have been known to intellectually isolate users from a variety of perspectives, or cause filter bubbles. In our work, we characterize and mitigate this filter bubble effect. We do so by classifying various datapoints based on their user-item interaction history and calculating the influences of the classified categories on each other using the well known TracIn method. Finally, we mitigate this filter bubble effect without compromising accuracy by carefully retraining our recommender system.