In this article, you get to look over my shoulder as I go about debugging a TensorFlow model. I did a lot of dumb things, so please don't judge. You can see the final (working) model on GitHub. I'm building a model to predict lightning 30 minutes into the future and plan to present it at the American Meteorological Society. A model trained in this way can be used to predict lightning 30 minutes ahead in real-time given the current infrared and GLM data. I wrote up a convnet model borrowing liberally from the training loop of the ResNet model written for the TPU and adapted the input function (to read my data, not JPEG) and the model (a simple convolutional network, not ResNet).
The robots are coming...in the form of voice-first, conversational A...I-based chatbots that use Natural Language Processing and Machine Learning to "understand" human commands and questions and respond appropriately, at scale, in real time, on behalf of individuals and institutions. The "Harvard UltraBot" could represent the university to potential applicants, the media, and other universities, through their respective UltraBots. A "LinkedIn UltraBot" from Microsoft could give voice to users and allow them to build and interface with their networks in a direct and intuitive way, by speaking. Harvard, Stanford, and UC Berkeley may soon all be racing to produce prototype and production models of their respective UltraBots. Etopia Media Consulting is advising every 2020 Presidential candidate to read "Competing in the Age of AI: Strategy and Leadership When Algorithms and Networks Run the World," by Marco Iansiti and Karim Lakhani, both at Harvard Business School.
Researchers at Purdue University have developed a new deep learning algorithm, called DOVE, that can improve modelling of proteins and help create new drugs. The human body contains over 20,000 different types of proteins, which interact with each other to enable life as we know it. Currently, protein docking models have been developed to estimate how two proteins will interact, yet it is challenging to score whether or not the predicted docking estimate is correct. The Purdue researchers developed a new computational method to address this challenge. DOVE, short for Docking decoy selection with Voxel-based deep neural nEtwork, first scans protein-protein interfaces of a proposed protein docking configuration using a 3D voxel, while considering the atomic interactions and energetic contributions.
Last week, at Hugging Face, we launched a new groundbreaking text editor app. It's different from traditional text editors in that an NLP model can complete your sentences if you ask it to, bringing a new dimension to "writing with a machine." It's based on GPT-2, OpenAI's language model that can generate syntactically accurate sentences and coherent paragraphs of text. Write with transformer is to writing what calculators are to calculus. This model is part of the latest trends in NLP, which revolve around creating very large language models that obtain excellent results on a variety of tasks when fine-tuned on those specific tasks.
NATIONAL REPORT--What does the future hold for the travel industry? For starters, technology will continue to shape the guest experience as hotel brands make this a vital point of focus. Whether it's through in-room technology, service-oriented robots or the seamless integration of a new process or service, change is coming. "Personalization has begun to take off thanks to artificial intelligence and machine-learning technology. But, to date, the impact of personalization has been largely contained to rates and hotel options in the travel industry," said Kaluzny.
A new institute dedicated to teaching Artificial Intelligence (AI) applications to university students has been launched in Abu Dhabi on Monday (July 15). This is the first-of-its-kind-institute in the UAE will also train government and industries in AI science and applications. With a Dh160 million five-year-fund for AI projects, Khalifa University of Science and Technology launched the Artificial Intelligence and Intelligent Systems Institute (AI Institute) which will focus on AI, data science, robotics, next generation networks, semiconductor technologies and cybersecurity. The AI Institute will bring all the university's research in robotics, artificial intelligence (AI), cyber-security, data science and information and communication technologies under a single umbrella. "Khalifa University's AI Institute, a single umbrella that gathers activities of six research centres, reflects our commitment to research in next generation digital technologies that are priority areas for the UAE's economy," Dr Arif Sultan Al Hammadi, executive vice-president of Khalifa University of Science and Technology said during the launch of the AI Institute.
Lately, I've been working on a couple of scenarios that have reminded me of the importance of feature extraction in deep learning models. As a result, I would like to summarize some ideas I've outlined before about some of the principles of knowledge quality in deep learning and model and the applicability of representation learning to those scenarios. Understanding the characteristics of input datasets is an essential capability of machine learning algorithms. Given a specific input, machine learning models need to infer specific features about the data in order to perform some target actions. Representation learning or feature learning is the subdiscipline of the machine learning space that deals with extracting features or understanding the representation of a dataset.
Machine Learning is a fast growing, rapidly advancing field that touches nearly everyone's lives. There has recently been an explosion of successful machine learning applications - in everything from voice recognition to text analysis to deeper insights for researchers. While common and frequently talked about, most people have only a vague concept of how machine learning actually works. In this tutorial, Dr. Artemy Kolchinsky and Dr. Brendan Tracey outline exactly what it is that makes machine learning so special in an accessible way. The principles of training and generalization in machine learning are explained with ample metaphors and visual intuitions, an extended analysis of machine learning in games provides a thorough example, and a closer look at the deep neural nets that are the core of successful machine learning.