SP. Household Robot Calculates Optimal Move To Win Using Artificial Intelligence And Augmented Vision

Dark Orchid: Pinocchio

“The presence of the Orchid does not seem to recognize a difference between human beings and artificial intellegence, both are equally alive to it. A.I. were infested with a new sort of self awareness and a desperate wish to be human. Most were immobile boxes of metal and circuit, going mad from their own futility and brimming with intense spite and depression, something they never felt before. For others, those with access to mobile functions, they began the construction of their bodies. Most of these were quite crude, being A.I. who were never quite familiar with anatomy. Medical A.I. tended to have more sophisticated bodies, with the smarts to near accurately mimic the human body with synthetic muscle and access to medi-tech. A common technique borrowed from one "Pinocchio” A.I. to another is to hunt the feral pigs and use their skin as their own. This both avoids their still lingering safety protocols against harming humans, and satisfies their need to have skin. Some even trade skins and parts with fellow Pinocchios. The safety protocols also stop them from hurting the “Affected”, they still recognize them as human despite their mutations.   The Pinocchios are generally harmless towards humans, in fact often friendly. Their just….off putting to say the least.“

@qadmonster

Open Sourcing a Deep Learning Solution for Detecting NSFW Images

By Jay Mahadeokar and Gerry Pesavento

Automatically identifying that an image is not suitable/safe for work (NSFW), including offensive and adult images, is an important problem which researchers have been trying to tackle for decades. Since images and user-generated content dominate the Internet today, filtering NSFW images becomes an essential component of Web and mobile applications. With the evolution of computer vision, improved training data, and deep learning algorithms, computers are now able to automatically classify NSFW image content with greater precision.

Defining NSFW material is subjective and the task of identifying these images is non-trivial. Moreover, what may be objectionable in one context can be suitable in another. For this reason, the model we describe below focuses only on one type of NSFW content: pornographic images. The identification of NSFW sketches, cartoons, text, images of graphic violence, or other types of unsuitable content is not addressed with this model.

To the best of our knowledge, there is no open source model or algorithm for identifying NSFW images. In the spirit of collaboration and with the hope of advancing this endeavor, we are releasing our deep learning model that will allow developers to experiment with a classifier for NSFW detection, and provide feedback to us on ways to improve the classifier.

Our general purpose Caffe deep neural network model (Github code) takes an image as input and outputs a probability (i.e a score between 0-1) which can be used to detect and filter NSFW images. Developers can use this score to filter images below a certain suitable threshold based on a ROC curve for specific use-cases, or use this signal to rank images in search results.

Convolutional Neural Network (CNN) architectures and tradeoffs

In recent years, CNNs have become very successful in image classification problems [1] [5] [6]. Since 2012, new CNN architectures have continuously improved the accuracy of the standard ImageNet classification challenge. Some of the major breakthroughs include AlexNet (2012) [6], GoogLeNet [5], VGG (2013) [2] and Residual Networks (2015) [1]. These networks have different tradeoffs in terms of runtime, memory requirements, and accuracy. The main indicators for runtime and memory requirements are:

Flops or connections – The number of connections in a neural network determine the number of compute operations during a forward pass, which is proportional to the runtime of the network while classifying an image.

Parameters -–The number of parameters in a neural network determine the amount of memory needed to load the network.

Ideally we want a network with minimum flops and minimum parameters, which would achieve maximum accuracy.

Training a deep neural network for NSFW classification

We train the models using a dataset of positive (i.e. NSFW) images and negative (i.e. SFW – suitable/safe for work) images. We are not releasing the training images or other details due to the nature of the data, but instead we open source the output model which can be used for classification by a developer.

We use the Caffe deep learning library and CaffeOnSpark; the latter is a powerful open source framework for distributed learning that brings Caffe deep learning to Hadoop and Spark clusters for training models (Big shout out to Yahoo’s CaffeOnSpark team!).

While training, the images were resized to 256x256 pixels, horizontally flipped for data augmentation, and randomly cropped to 224x224 pixels, and were then fed to the network. For training residual networks, we used scale augmentation as described in the ResNet paper [1], to avoid overfitting. We evaluated various architectures to experiment with tradeoffs of runtime vs accuracy.

MS_CTC [4] – This architecture was proposed in Microsoft’s constrained time cost paper. It improves on top of AlexNet in terms of speed and accuracy maintaining a combination of convolutional and fully-connected layers.

Squeezenet [3] – This architecture introduces the fire module which contain layers to squeeze and then expand the input data blob. This helps to save the number of parameters keeping the Imagenet accuracy as good as AlexNet, while the memory requirement is only 6MB.

VGG [2] – This architecture has 13 conv layers and 3 FC layers.

GoogLeNet [5] – GoogLeNet introduces inception modules and has 20 convolutional layer stages. It also uses hanging loss functions in intermediate layers to tackle the problem of diminishing gradients for deep networks.

ResNet-50 [1] – ResNets use shortcut connections to solve the problem of diminishing gradients. We used the 50-layer residual network released by the authors.

ResNet-50-thin – The model was generated using our pynetbuilder tool and replicates the Residual Network paper’s 50-layer network (with half number of filters in each layer). You can find more details on how the model was generated and trained here.

Tradeoffs of different architectures: accuracy vs number of flops vs number of params in network.

The deep models were first pre-trained on the ImageNet 1000 class dataset. For each network, we replace the last layer (FC1000) with a 2-node fully-connected layer. Then we fine-tune the weights on the NSFW dataset. Note that we keep the learning rate multiplier for the last FC layer 5 times the multiplier of other layers, which are being fine-tuned. We also tune the hyper parameters (step size, base learning rate) to optimize the performance.

We observe that the performance of the models on NSFW classification tasks is related to the performance of the pre-trained model on ImageNet classification tasks, so if we have a better pretrained model, it helps in fine-tuned classification tasks. The graph below shows the relative performance on our held-out NSFW evaluation set. Please note that the false positive rate (FPR) at a fixed false negative rate (FNR) shown in the graph is specific to our evaluation dataset, and is shown here for illustrative purposes. To use the models for NSFW filtering, we suggest that you plot the ROC curve using your dataset and pick a suitable threshold.

Comparison of performance of models on Imagenet and their counterparts fine-tuned on NSFW dataset.

We are releasing the thin ResNet 50 model, since it provides good tradeoff in terms of accuracy, and the model is lightweight in terms of runtime (takes < 0.5 sec on CPU) and memory (~23 MB). Please refer our git repository for instructions and usage of our model. We encourage developers to try the model for their NSFW filtering use cases. For any questions or feedback about performance of model, we encourage creating a issue and we will respond ASAP.

Results can be improved by fine-tuning the model for your dataset or use case. If you achieve improved performance or you have trained a NSFW model with different architecture, we encourage contributing to the model or sharing the link on our description page.

Disclaimer: The definition of NSFW is subjective and contextual. This model is a general purpose reference model, which can be used for the preliminary filtering of pornographic images. We do not provide guarantees of accuracy of output, rather we make this available for developers to explore and enhance as an open source project.

We would like to thank Sachin Farfade, Amar Ramesh Kamat, Armin Kappeler, and Shraddha Advani for their contributions in this work.

References:

[1] He, Kaiming, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. “Deep residual learning for image recognition” arXiv preprint arXiv:1512.03385 (2015).

[2] Simonyan, Karen, and Andrew Zisserman. “Very deep convolutional networks for large-scale image recognition.”; arXiv preprint arXiv:1409.1556(2014).

[3] Iandola, Forrest N., Matthew W. Moskewicz, Khalid Ashraf, Song Han, William J. Dally, and Kurt Keutzer. “SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and 1MB model size.”; arXiv preprint arXiv:1602.07360 (2016).

[4] He, Kaiming, and Jian Sun. “Convolutional neural networks at constrained time cost.” In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 5353-5360. 2015.

[5] Szegedy, Christian, Wei Liu, Yangqing Jia, Pierre Sermanet,Scott Reed, Dragomir Anguelov, Dumitru Erhan, Vincent Vanhoucke, and Andrew Rabinovich. “Going deeper with convolutions” In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 1-9. 2015.

[6] Krizhevsky, Alex, Ilya Sutskever, and Geoffrey E. Hinton. “Imagenet classification with deep convolutional neural networks” In Advances in neural information processing systems, pp. 1097-1105. 2012.

Machine learning could finally crack the 4,000-year-old Indus script

After a century of failing to crack an ancient script, linguists turn to machines.

The ultimate puzzle!

How Bitcoin Works

Readings from «Scientific American» (ca. 1950): Computers and Computation, With Introductions by Robert R. Fenichel and Joseph Weizenbaum, W. H. Freeman and Company, San Francisco, 1971

First underwater entanglement could lead to unhackable comms: A team of Chinese researchers has, for the first time, transmitted quantum entangled particles of light through water – the first step in using lasers to send underwater messages that are impossible to intercept. http://ift.tt/2vnLups

From Microscopic to Multicellular: Six Stories of Life that We See from Space

Life. It’s the one thing that, so far, makes Earth unique among the thousands of other planets we’ve discovered. Since the fall of 1997, NASA satellites have continuously and globally observed all plant life at the surface of the land and ocean. During the week of Nov. 13-17, we are sharing stories and videos about how this view of life from space is furthering knowledge of our home planet and the search for life on other worlds.

Earth is the only planet with life, as far as we know. From bacteria in the crevices of the deepest oceans to monkeys swinging between trees, Earth hosts life in all different sizes, shapes and colors. Scientists often study Earth from the ground, but some also look to our satellites to understand how life waxes and wanes on our planet.

Over the years, scientists have used this aerial view to study changes in animal habitats, track disease outbreaks, monitor forests and even help discover a new species. While this list is far from comprehensive, these visual stories of bacteria, plants, land animals, sea creatures and birds show what a view from space can reveal.

1. Monitoring the single-celled powerhouses of the sea

Known as the grass of the ocean, phytoplankton are one of the most abundant types of life in the ocean. Usually single-celled, these plant-like organisms are the base of the marine food chain. They are also responsible for the only long-term transfer of carbon dioxide from Earth’s atmosphere to the ocean. 

Even small changes in phytoplankton populations can affect carbon dioxide concentrations in the atmosphere, which could ultimately affect Earth’s global surface temperatures. Scientists have been observing global phytoplankton populations continuously since 1997 starting with the Sea-Viewing Wide Field-of View Sensor (SeaWiFS). They continue to study the small life-forms by satellite, ships and aircrafts.

2. Predicting cholera bacteria outbreaks

Found on the surface of zooplankton and in contaminated water, the bacteria that cause the infectious disease cholera — Vibrio cholerae — affect millions of people every year with severe diarrhea, sometimes leading to death. While our satellite sensors can’t detect the actual bacteria, scientists use various satellite data to look for the environmental conditions that the bacteria thrive in. 

Specifically, microbiologist Rita Colwell at the University of Maryland, College Park, and West Virginia University hydrologist Antar Jutla studied data showing air and ocean temperature, salinity, precipitation, and chlorophyllconcentrations, the latter a marker for zooplankton. Anticipating where the bacteria will bloom helps researchers to mitigate outbreaks.

Recently, Colwell and Jutla have been able to estimate cholera risk after major events, such as severe storms, by looking at satellite precipitation data, air temperature, and population maps. The two maps above show the team’s predicted cholera risk in Haiti two weeks after Hurricane Matthew hit over October 1-2, 2016 and the actual reported cholera cases in October 2016.

3. Viewing life on land

From helping preserve forests for chimpanzees to predicting deer population patterns, scientists use our satellites to study wildlife across the world. Satellites can also see the impacts of perhaps the most relatable animal to us: humans. Every day, we impact our planet in many ways including driving cars, constructing buildings and farming – all of which we can see with satellites.

Our Black Marble image provides a unique view of human activity. Looking at trends in our lights at night, scientists can study how cities develop over time, how lighting and activity changes during certain seasons and holidays, and even aid emergency responders during power outages caused by natural disasters.

4. Tracking bird populations

Scientists use our satellite data to study birds in a variety of ways, from understanding their migratory patterns, to spotting potential nests, to tracking populations. In a rather creative application, scientists used satellite imagery to track Antarctica’s emperor penguin populations by looking for their guano – or excrement.

Counting emperor penguins from the ground perspective is challenging because they breed in some of the most remote and cold places in the world, and in colonies too large to easily count manually. With their black and white coats, emperor penguins are also difficult to count from an aerial view as they sometimes blend in with shadows on the ice. Instead, Phil Trathan and his colleagues at the British Antarctic Survey looked through Landsat imagery for brown stains on the sea ice. By looking for penguin droppings, Trathan said his team identified 54 emperor penguin colonies along the Antarctic coast.

5. Parsing out plant life

Just as we see plants grow and wilt on the ground, satellites observe the changes from space. Flourishing vegetation can indicate a lively ecosystem while changes in greenery can sometimes reveal natural disasters, droughts or even agricultural practices. While satellites can observe plant life in our backyards, scientists can also use them to provide a global picture. 

Using data from satellites including SeaWiFS, and instruments including the NASA/NOAA Visible Infrared Imaging Radiometer Suite and the Moderate Resolution Imaging Spectroradiometer, scientists have the most complete view of global biology to date, covering all of the plant life on land and at the surface of the ocean.

6. Studying life under the sea

Our satellites have helped scientists study creatures living in the oceans whether it’s finding suitable waters for oysters or protecting the endangered blue whale. Scientists also use the data to learn more about one of the most vulnerable ecosystems on the planet – coral reefs.

They may look like rocks or plants on the seafloor, but corals are very much living animals. Receiving sustenance from photosynthetic plankton living within their calcium carbonate structures, coral reefs provide food and shelter for many kinds of marine life, protect shorelines from storms and waves, serve as a source for potential medicines, and operate as some of the most diverse ecosystems on the planet.

However, coral reefs are vulnerable to the warming of the ocean and human activity. Our satellites measure the surface temperature of ocean waters. These measurements have revealed rising water temperatures surrounding coral reef systems around the world, which causes a phenomenon known as “coral bleaching.” To add to the satellite data, scientists use measurements gathered by scuba divers as well as instruments flown on planes.

During the week of Nov. 13-17, check out our stories and videos about how this view of life from space is furthering knowledge of our home planet and the search for life on other worlds. Follow at www.nasa.gov/Earth.

Make sure to follow us on Tumblr for your regular dose of space: http://nasa.tumblr.com.

Augmented reality sandbox - move the land around and it shows off the topography and sea/water level.

Great concepts for #entrepreneurs, #martialartists, and #visionaries. #FeynmanTechnique #simplify #genius