Oleg Zabluda's blog
Tuesday, September 27, 2016
 
Trump vs. Hillary: Debate Battle Rap
Trump vs. Hillary: Debate Battle Rap
https://www.youtube.com/watch?v=U2Zxc4WuJ6Q
https://www.youtube.com/watch?v=U2Zxc4WuJ6Q

Labels:


 
Sick Berns: Hillary Debates Hillary
Sick Berns: Hillary Debates Hillary
https://www.youtube.com/watch?v=eulMm9LVg00
https://www.youtube.com/watch?v=eulMm9LVg00

Labels:


 
Purge the Bigots, Brendan Eich is just the beginning.
Purge the Bigots, Brendan Eich is just the beginning. Let’s oust everyone who donated to the campaign against gay marriage
"""
More than 35,000 people gave money to the campaign for Proposition 8, the 2008 ballot measure that declared, “Only marriage between a man and a woman is valid or recognized in California.” [...] To organize the next stage of the purge, I’ve compiled the financial data into three tables.
"""
http://www.slate.com/articles/news_and_politics/frame_game/2014/04/brendan_eich_quits_mozilla_let_s_purge_all_the_antigay_donors_to_prop_8.html
http://www.slate.com/articles/news_and_politics/frame_game/2014/04/brendan_eich_quits_mozilla_let_s_purge_all_the_antigay_donors_to_prop_8.html

Labels:


 
European Foundation for Support of Culture on “Matinee Concert”
European Foundation for Support of Culture on “Matinee Concert”
"""
“Matinee Concert” of the famous German pianist Albert Mamriev was held at spectacular Hall Wolkenburg in Cologne on the 25th of September 2015 organized by European Foundation for Support of Culture (President Konstantin Ishkhanov).
[...]
This piano recital was dedicated to the honorable modern composer Alexey Shor . Alexey Shor’s works have been performed in diverse locales around the world by, among others, the Orchestre National des Pays de la Loire, the Hamburg Opera Orchestra, the George Enescu Philharmonic Orchestra, the Slovak Sinfonietta and the “Soloists of Russia.” His compositions have been featured in Schloss Elmau’s music series in the Bavarian Alps of Germany, the Verbier Festival in Switzerland, the Aldeburgh Festival in the United Kingdom, at Carnegie Hall and the Metropolitan Museum of Art in New York.

In the performance of Albert Mamriev the audience heard stunning, exciting and romantic works from the "Childhood Memories" Suite and the "European notebook" Suite by Alexey Shor. These works were greeted with a big enthusiasm – the audience was watching and admiring the tremendous professional recital with pleasure that interested and inspired with every note.
"""
https://www.facebook.com/eufsc.eu/posts/1159596944134655
https://www.facebook.com/alexey.shor.1/posts/530901550446048



Labels:


 
A Neural Network for Machine Translation, at Production Scale
A Neural Network for Machine Translation, at Production Scale
"""
Ten years ago, we announced the launch of Google Translate, together with the use of Phrase-Based Machine Translation as the key algorithm behind this service. [...] Today we announce the Google Neural Machine Translation system (GNMT), [...] A few years ago we started using Recurrent Neural Networks (RNNs) to directly learn the mapping between an input sequence (e.g. a sentence in one language) to an output sequence (that same sentence in another language) [2]. Whereas Phrase-Based Machine Translation (PBMT) breaks an input sentence into words and phrases to be translated largely independently, Neural Machine Translation (NMT) considers the entire input sentence as a unit for translation. [...] Since then, researchers have proposed many techniques to improve NMT, including work on handling rare words by mimicking an external alignment model [3], using attention to align input words and output words [4] and breaking words into smaller units to cope with rare words [5,6]. Despite these improvements, NMT wasn't fast or accurate enough to be used in a production system, such as Google Translate. Our new paper [1] describes how we overcame the many challenges to make NMT work on very large data sets and built a system that is sufficiently fast and accurate enough to provide better translations for Google’s users and services.
[...]
compared to the previous phrase-based production system. GNMT reduces translation errors by more than 55%-85% on several major language pairs measured on sampled sentences from Wikipedia and news websites with the help of bilingual human raters. [...] GNMT can still make significant errors that a human translator would never make, like dropping words and mistranslating proper names or rare terms, and translating sentences in isolation rather than considering the context of the paragraph or page.
"""
https://research.googleblog.com/2016/09/a-neural-network-for-machine.html

[1] Google’s Neural Machine Translation System: Bridging the Gap between Human and Machine Translation (2016) Yonghui Wu, [...] Quoc V. Le, [...] Oriol Vinyals, Greg Corrado, [...] Jeffrey Dean.

[2] Sequence to Sequence Learning with Neural Networks (2014) Ilya Sutskever, Oriol Vinyals, Quoc V. Le.

Labels:


 
You Only Look Once: Unified, Real-Time Object Detection (2016) Joseph Redmon, [...] Ross Girshick et al
You Only Look Once: Unified, Real-Time Object Detection (2016) Joseph Redmon, [...] Ross Girshick et al
"""
Prior work on object detection repurposes classifiers to perform detection. Instead, we frame object detection as a regression problem to spatially separated bounding boxes and associated class probabilities. [...] can be optimized end-to-end directly on detection performance. [...] extremely fast. [...] 45 frames per second. A smaller version of the network, Fast YOLO, processes an astounding 155 frames per second while still achieving double the mAP of other real-time detectors. Compared to state-of-the-art detection systems, YOLO makes more localization errors but is far less likely to predict false detections where nothing exists. Finally, YOLO learns very general representations of objects. It outperforms all other detection methods, including DPM and R-CNN, by a wide margin when generalizing from natural images to artwork on both the Picasso Dataset and the People-Art Dataset.
[...]
Table 1: Real-Time Systems on PASCAL VOC 2007. Comparing
the performance and speed of fast detectors.

R-CNN minus R replaces Selective Search with static bounding box proposals [20]. While it is much faster than R-CNN, it still falls short of real-time and takes a significant accuracy hit from not having good proposals. Fast R-CNN speeds up the classification stage of R-CNN but it still relies on selective search which can take around 2 seconds per image to generate bounding box proposals. Thus it has high mAP but at 0.5 fps it is still far from realtime. The recent Faster R-CNN replaces selective search with a neural network to propose bounding boxes, similar to Szegedy et al. [8] In our tests, their most accurate model achieves 7 fps while a smaller, less accurate one runs at 18 fps. The VGG-16 version of Faster R-CNN is 10 mAP higher but is also 6 times slower than YOLO. The ZeilerFergus Faster R-CNN is only 2.5 times slower than YOLO but is also less accurate.
"""
https://arxiv.org/abs/1506.02640
https://arxiv.org/abs/1506.02640

Labels:



Powered by Blogger