Collaborating Authors


Understanding GANs


GANs (Generative Adversarial Networks) are a class of models where images are translated from one distribution to another. GANs are helpful in various use-cases, for example: enhancing image quality, photograph editing, image-to-image translation, clothing translation, etc. Nowadays, many retailers, fashion industries, media, etc. are making use of GANs to improve their business and relying on algorithms to do the task. There are many forms of GAN available serving different purposes, but in this article, we will focus on CycleGAN. Here we will see its working and implementation in PyTorch. CycleGAN learns the mapping of an image from source X to a target domain Y. Assume you have an aerial image of a city and want to convert in google maps image or the landscape image into a segmented image, but you don't have the paired images available, then there is GAN for you.

Selene Desiré Bovo posted on LinkedIn


How long will MT need humans? "Translation involves many things that don't fit common definitions." But it still faces some issues, such as sentence structure, shades of meaning, cultural context and the use of language. "Language relies on intent, on shared secrets, on group identity and on hidden knowledge." We are constantly creating new language.

Hot papers on arXiv from the past month – July 2020


Here are the most tweeted papers that were uploaded onto arXiv during July 2020. Results are powered by Arxiv Sanity Preserver. Abstract: Massive language models are the core of modern NLP modeling and have been shown to encode impressive amounts of commonsense and factual information. However, that knowledge exists only within the latent parameters of the model, inaccessible to inspection and interpretation, and even worse, factual information memorized from the training corpora is likely to become stale as the world changes. Knowledge stored as parameters will also inevitably exhibit all of the biases inherent in the source materials.

OpenAI's new GPT-3 language explained in under 3 minutes


So, you've seen some amazing GPT-3 demos on Twitter (if not, where have you been?). This mega machine learning model, created by OpenAI, can write it's own op-eds, poems, articles, and even working code: With GPT-3, I built a layout generator where you just describe any layout you want, and it generates the JSX code for you. GPT3()… the spreadsheet function to rule them all. Impressed with how well it pattern matches from a few examples. The same function looked up state populations, peoples' twitter usernames and employers, and did some math.

Semantic interoperability in IoT


Semantic interoperability includes the ability to establish a shared meaning of the data exchanged, as well as the ability to similarly interpret communication interfaces. Shared meaning here means that two different computer systems, for example, not only can communicate data in the basic sense (such as an integer with value 42), but also attach unambiguous meaning to the data. For example, radiator three's temperature in the conference room on level five is currently 42 Celsius. As we build large IoT systems we are faced with several challenges of scale. Among them is the challenge of being able to make equipment and subsystems of different vendors interoperable and, over different time periods, work together and as intended.

The Transformer Isn't As Hard To Understand As You Might Think


Transformer's architecture has been the cornerstone for the development of many of the latest SOTA NLP models. It mainly relies on a mechanism called attention. Unlike other successful models that came before, it has no involvement with convolutional or recurrent layers what so ever. If you're new to this model, chances are you won't find this architecture to be easiest to understand. If that's the case, I hope this article can help.

How Applications of Deep Learning Revolutionizing Industries?


Deep learning, a subset of artificial intelligence, is already making its way into day-to-day aspects of life and business. A few years back, the technology was touted to be the futuristic concept as it differs from traditional machine learning systems. Today, deep learning is capable of self-learning and improving as it assesses large data sets. It has a large number of business applications and has the potential to revolutionize industries, emerging as the next big disruption of AI. Deep learning is typically designed to imitate the way the human brain processes data.

Deep learning to translate between programming languages


Migrating a codebase from an archaic programming language such as COBOL to a modern alternative like Java or C is a difficult, resource-intensive task that requires expertise in both the source and target languages. COBOL, for example, is still widely used today in mainframe systems around the world, so companies, governments, and others often must choose whether to manually translate their code bases or commit to maintaining code written in a language that dates back to the 1950s. We've developed TransCoder, an entirely self-supervised neural transcompiler system that can make code migration far easier and more efficient. Our method is the first AI system able to translate code from one programming language to another without requiring parallel data for training. We've demonstrated that TransCoder can successfully translate functions between C, Java, and Python 3. TransCoder outperforms open source and commercial rule-based translation programs.

CoVoST V2: Expanding the largest, most diverse multilingual speech-to-text translation data set


CoVoST V2 expands on our CoVoST data set, a speech-to-text translation (ST) corpus targeted at multilingual translation. This new release makes available the largest multilingual ST data set to date. CoVoST V2 will facilitate translating 21 languages into English, as well as English into 15 languages. In order to support wider research and applications in multilingual speech translation, we have released CoVoST V2 as free to use via a Creative Commons (CC0) license. Developed in 2019, the initial version of CoVoST used Mozilla's open source Common Voice database of crowdsourced voice recordings to create a corpus for translating 11 languages into English, with diverse speakers and accents.

Leaders from Google, Adobe, and more talk benefits and bias at the Conversational AI Summit


"I'm extremely excited about the future of the intersection between conversational AI and the multitude of platforms that are being developed around these capabilities," said Linden Hillebrand, VP Global Customer Success and Support at Cloudera during his opening remarks at the Transform 2020 Conversational AI Summit. Over the course of the day tech giants from Adobe and Capital One to Google, Amazon, and Twitter spoke about how they're using conversational AI to solve problems for their businesses in new and innovative ways. The technology is being leveraged for both text chatbots and the NLP-powered voice assistants that are increasingly able to understand intent and offer a seamless, personalized user experience, helping automate the majority of customer interactions. But in most sessions, panelists emphasized that implementing these AI technologies also means tackling some of the bigger picture issues, including fairness, explainability, and elimination of bias. Data company Cloudera had a head start in developing a conversational AI platform: the vast data sets they had stored from past customer issues and solutions.