Lifeboat Foundation

A team of researchers from OpenAI recently published a paper describing GPT-3, a deep-learning model for natural-language with 175 billion parameters, 100x more than the previous version, GPT-2. The model is pre-trained on nearly half a trillion words and achieves state-of-the-art performance on several NLP benchmarks without fine-tuning.

In paper published on arXiv, a team of over 30 co-authors described the model and several experiments. The researchers’ goal was to produce an NLP system that performs well on a variety of tasks with little or no fine-tuning, and previous work had indicated that larger models might be the solution. To test that hypothesis, the team increased the size of their previous model, GPT-2, from 1.5 billion parameters to 175 billion. For training, the team collected several datasets, including the Common Crawl dataset and the English-language Wikipedia. The model was evaluated against several NLP benchmarks, matching state-of-the-art performance on “closed-book” question-answering tasks and setting a new record for the LAMBADA language modeling task.

OpenAI made headlines last year with GPT-2 and their decision not to release the 1.5 billion parameter version of the trained model due to “concerns about malicious applications of the technology.” GPT-2 is one of many large-scale NLP models based on the Transformer architecture. These models are pre-trained on large text corpora, such as the contents Wikipedia, using self-supervised learning. In this scenario, instead of using a dataset containing inputs paired with expected outputs, the model is given a sequence of text with words “masked” and it must learn to predict the masked words based on the surrounding context. After this pre-training, the models are then fine-tuned with a labelled benchmark dataset for a particular NLP task, such as question-answering.

DARPA’s hypersonic glider sustained controlled flight at Mach 20 for three minutes.

Authorities are testing a new kind of money in four cities.

As the pandemic’s economic toll grows around the world, some experts fear it could harm science for decades by putting many thousands of researchers out of work and forcing nations to slash funding as they rebuild societies. Others say the pandemic could highlight the importance of science and spur long-term support, especially for basic research, much as the Second World War did.

Financial crises could spell trouble for science budgets but spending could surge in some countries: part 2 in a series on science after the pandemic.

Skateboarding legend Rodney Mullen teams up with Physics Girl to explain the unusual physics behind skateboard tricks. Filmed with a phantom high speed camera at 1000fps, see Mullen’s tricks like never before.

If you liked this video check out these:
How SMOOTHNESS of a SOCCER BALL affects curve!

Crazy tic tac bounce!?

Continue reading “Why this trick should be IMPOSSIBLE ft. Rodney Mullen — Skateboarding Science” »

For more than six decades, the quest to understand the formation of hot (about 20,000−30,000 K) extreme horizontal branch (EHB) stars in Galactic globular clusters has remained one of the most elusive in stellar evolutionary theory. Here we report on two discoveries that challenge the idea of the stable luminosity of EHB stars. The first mode of EHB variability is periodic and cannot be ascribed to either binary evolution or pulsation. Instead, we attribute it here to the presence of magnetic spots: superficial chemical inhomogeneities whose projected rotation induces the variability. The second mode of EHB variability is aperiodic and manifests itself on timescales of years. In two cases, six-year-long light curves display superflare events that are several million times more energetic than solar analogues. We advocate a scenario in which the two EHB variability phenomena are different manifestations of diffuse, dynamo-generated, weak magnetic fields. Magnetism is therefore a key player driving the formation and evolution of EHB clusters stars and, likewise, operating in the Galactic field counterparts. Our conclusions bridge similar variability/magnetism phenomena in all radiative-enveloped hot-stars: young main-sequence stars, old EHBs and defunct white dwarfs.

Researchers are trying to store robust quantum information in Majorana particles and are generating quantum gates by exploiting the bizarre non-abelian statistics of Majorana zero modes bound to topological defects.

Government agencies and universities around the world—not to mention tech giants like IBM and Google—are vying to be the first to answer a trillion-dollar quantum question : How can quantum computers reach their vast potential when they are still unable to consistently produce results that are reliable and free of errors?

Every aspect of these exotic machines—including their fragility and engineering complexity; their preposterously sterile, low-temperature operating environment; complicated mathematics; and their notoriously shy quantum bits (qubits) that flip if an operator so much as winks at them—are all potential sources of errors. It says much for the ingenuity of scientists and engineers that they have found ways to detect and correct these errors and have quantum computers working to the extent that they do: at least long enough to produce limited results before errors accumulate and quantum decoherence of the qubits kicks in.

This paper describes the design, implementation, and evaluation of VanarSena, an automated fault finder for mobile applications (“apps’‘). The techniques in VanarSena are driven by a study of 25 million real-world crash reports of Windows Phone apps reported in 2012. Our analysis indicates that a modest number of root causes are responsible for many observed failures, but that they occur in a wide range of places in an app, requiring a wide coverage of possible execution paths. VanarSena adopts a “greybox’’ testing method, instrumenting the app binary to achieve both coverage and speed. VanarSena runs on cloud servers: the developer uploads the app binary; VanarSena then runs several app “monkeys’’ in parallel to emulate user, network, and sensor data behavior, returning a detailed report of crashes and failures. We have tested VanarSena with 3000 apps from the Windows Phone store, finding that 1108 of them had failures; VanarSena uncovered 2969 distinct bugs in existing apps, including 1227 that were not previously reported. Because we anticipate VanarSena being used in regular regression tests, testing speed is important. VanarSena uses two techniques to improve speed. First, it uses a “hit testing’’ method to quickly emulate an app by identifying which user interface controls map to the same execution handlers in the code. Second, it generates a ProcessingCompleted event to accurately determine when to start the next interaction. These features are key benefits of VanarSena’s greybox philosophy.

2014-06

http://hdl.handle.net/1721.1/110759