Collaborating Authors

[R] Generative Adversarial Transformers (2103.01209)


Abstract: We introduce the GANsformer, a novel and efficient type of transformer, and explore it for the task of visual generative modeling. The network employs a bipartite structure that enables long-range interactions across the image, while maintaining computation of linearly efficiency, that can readily scale to high-resolution synthesis. It iteratively propagates information from a set of latent variables to the evolving visual features and vice versa, to support the refinement of each in light of the other and encourage the emergence of compositional representations of objects and scenes. In contrast to the classic transformer architecture, it utilizes multiplicative integration that allows flexible region-based modulation, and can thus be seen as a generalization of the successful StyleGAN network. We demonstrate the model's strength and robustness through a careful evaluation over a range of datasets, from simulated multi-object environments to rich real-world indoor and outdoor scenes, showing it achieves state-of-the-art results in terms of image quality and diversity, while enjoying fast learning and better data-efficiency.

Insights Discovery in Data Science Through Novel Machine Learning Approaches


I have always appreciated the unusual, unexpected, and surprising in science and in data. As famous science author Arthur C. Clarke once said, "The most exciting phrase to hear in science, the one that heralds new discoveries, is not'Eureka!' (I found it) but'That's funny!'" This is the primary reason that I motivated most of the doctoral students that I mentored at GMU to work on some variation of Novelty Discovery (or Surprise Discovery) for their Ph.D. dissertations. "Surprise discovery" for me is a much more positive, exciting phrase than "outlier detection" or "anomaly detection", and it is much richer in meaning, in algorithms, and in new opportunities. Finding the surprising unexpected thing in your data is what inspires our exclamation "That's funny!" that may be signaling a great discovery (either about your data's quality, or about your data pipeline's deficiencies, or about some wholly new scientific concept). As famous astronomer, Vera Rubin said, "Science progresses best when observations force us to alter our preconceptions."

Council Post: Leveraging AI And NLP For Automated Resolution Of Tasks


Pat Calhoun, a visionary leader focused on UX and adoption, is the CEO and Founder of Espressive, transforming enterprise self-help with AI. Enterprises are quickly shifting their IT help desk strategies away from one where every employee's issue or request requires human intervention to one that leverages artificial intelligence (AI)/natural language processing (NLP) for automated resolution. These are initial help desk automation platforms focused on providing automated responses to incidents or inquiries. However, as enterprises saw the value associated with reducing their dependency on humans in problem resolution, they started looking at what to automate next. One area that the enterprise service management (ESM) market is now focusing on is the automation of tasks (e.g., fulfill a service request, create a new mailing list, schedule PTO, reserve guest desk).

The modern hurdles to widespread AI adoption


Artificial Intelligence is used to inform and shape strategies across a range of industries, but there are still several challenges holding it back from widespread adoption. Ethical considerations must be addressed and operational difficulties, such as building a team with the right skill set, always provide an obstacle. COVID-19 has given organisations across the world the need to expand their digital services. At first glance this would appear to benefit the spread of machine learning. When more people move their financial transactions and activity online, there is more data to tally and learn from.

Dooly raises $20 million to organize sales information with AI


Vancouver, Canada-based Dooly, a startup developing an AI-powered plugin for customer relationship management (CMR) platforms, today announced that it raised $20 million, a combination of $3.3 million seed and $17 million series A tranches. The company plans to use the capital to scale its platform well into this year, according to CEO Kris Hartvigsen. Salesforce's 2019 State of Sales Report found that, on average, salespeople only spend 34% of their day selling products. Among the biggest culprits of the lost time is the disconnect between enterprises' need for a CRM and the fact that these platforms don't always map to how salespeople work. According to a recent survey, one of the top barriers to CRM adoption is the amount of manual data entry.

AI in Marketing and Sales for 2021 - What is beneficial for company growth


AI Implementations in Marketing and Sales Underwent a Hype in the Last Year, Now It Is a Norm to Seek Digital Changes for Efficiency and Acceleration. AI has been in our lives for a while, but its applications had been quite less before 2020 as compared to now. There were people who feared that bringing AI into the picture would trigger human replacement. Contrarily, there were also some who thought that AI would just take all the workload off their backs. In the following year, the doubts cleared off till a pretty good extent.

How Artificial Intelligence and Machine Learning are enhancing the learning curve for students


The advanced with regard to Artificial Intelligence and Machine Learning is also helping the students when it comes to content. A huge bank of data means a plethora of study material to choose from. With the help of AI & ML, multiple checkpoints are put in place, which evaluates the content on the basis of a few defined parameters to make relevant recommendations. This enables students to get shortlisted content material as per their specific aptitude. The role of User Generated Content cannot, however, be denied here.

Artificial Intelligence Has Yet to Fully Infiltrate Online Gaming


Original video games of the 1970s contained very little, if any, Artificial Intelligence (AI). Game code in these early days was made up of rather complex "if" statements that allowed for a fixed (and not always spontaneous) number of game choices and scenarios. Today's video games work using the same fundamental concepts that games created in the early 1980s and 1990s used; they're just scaled with more data and more processing power. That's not to say that the games themselves have not changed since 1982. Today's games have extraordinary graphics, sound, and stories compared to earlier trailblazers.

Build your own voice assistant with this DIY kit


With this coding kit, which is ideal for ages 11 and up, you can spend quality time with your kiddos creating Spencer, the DIY voice assistant. You'll follow step-by-step instructions to build Spencer. Don't worry, no prior coding or electronics knowledge is necessary. Along the way, you'll learn about coding a microcomputer in C, C, and CircuitBlocks (similar to Scratch), artificial intelligence, voice recognition, sound processing, soldering, and more. Once he's up and running, Spencer can tell you the weather forecast, sing a song, set alarms and reminders, show animations and scrolling text with flashy LEDs, find news online, and even tell jokes.

AIhub monthly digest: February 2021


Welcome to the second of our monthly digests, designed to keep you up-to-date with the happenings in the AI world. You can catch up with any AIhub stories you may have missed, get the low-down on recent conferences, and generally immerse yourself in all things AI. You may be aware that we are running a focus series on the UN sustainable development goals (SDG). Each month we tackle a different SDG and cover some of the AI research linked to that particular goal. In February it was the turn of climate action.