Building AI: 3 theorems you need to know – DXC Blogs

Building #AI: 3 theorems you need to know #MachineLearning

  • The mathematical theorem proving this is the so-called “no-free-lunch theorem” It tells us that if a learning algorithm works well with one kind of data, it will work poorly with other types of data.
  • In a way, a machine learning algorithm projects its own knowledge onto data.
  • In machine learning, overfitting occurs when your model performs well on training data, but the performance becomes horrible when switched to test data.
  • Any learning algorithm must also be a good model of the data; if it learns one type of data effectively, it will necessarily be a poor model — and a poor student – of some other types of data.
  • Good regulator theorem also tells us that determining if inductive bias will be beneficial or detrimental for modeling certain data depends on whether the equations defining the bias constitute a good or poor model of the data.

Editor’s note: This is a series of blog posts on the topic of “Demystifying the creation of intelligent machines: How does one create AI?” You are now reading part 3. For the list of all, see here: 1, 2, 3, 4, 5, 6, 7.
Continue reading “Building AI: 3 theorems you need to know – DXC Blogs”

Nightmare Hellface Generator is Cutting-Edge Machine Learning

Nightmare hellface generator is cutting-edge machine learning:

  • Draw something in a little box and an algorithm will try to interpret it as a cat and then fill in the colors and textures according to a machine learning model training on thousands of cat images.
  • The pix2pix project demonstrates something pretty profound about machine learning circa 2017: It’s awful at generating new images, or at least meaningful new images.
  • Machine learning is better at classifying existing images, but, even then, things drop off dramatically as we move beyond a handful of really robust object-recognition models.
  • GANs work by training generative models that seek to minimize a particular “loss function” according to a prediction that the generated image is fake or real.
  • Rather than learn how to produce images from scratch, the model here learns to map the abstract image representation contained within a machine learning model to a trackpad doodle.

Generative adversarial networks strike again.
Continue reading “Nightmare Hellface Generator is Cutting-Edge Machine Learning”

Nightmare Hellface Generator is Cutting-Edge Machine Learning

Nightmare hellface generator is cutting-edge machine learning

  • Draw something in a little box and an algorithm will try to interpret it as a cat and then fill in the colors and textures according to a machine learning model training on thousands of cat images.
  • The pix2pix project demonstrates something pretty profound about machine learning circa 2017: It’s awful at generating new images, or at least meaningful new images.
  • Machine learning is better at classifying existing images, but, even then, things drop off dramatically as we move beyond a handful of really robust object-recognition models.
  • GANs work by training generative models that seek to minimize a particular “loss function” according to a prediction that the generated image is fake or real.
  • Rather than learn how to produce images from scratch, the model here learns to map the abstract image representation contained within a machine learning model to a trackpad doodle.

Generative adversarial networks strike again.
Continue reading “Nightmare Hellface Generator is Cutting-Edge Machine Learning”

AI can now predict whether or not humans will think your photo is awesome

#AI can now predict whether or not humans will think your photo is awesome

  • The Aesthetics tool, still in beta testing, allows users to upload a photo and get an auto-generated list of tags, as well as a percentage rate on the “chance that this image is awesome.”
  • According to developers, the neural network was trained to view an image much in the same way a human photo editor would, looking at factors such as color, sharpness, and subject.
  • As early users report, the system seems to be fairly good at recognizing factors like whether or not the image is sharp and if the composition is interesting, but it is certainly far from a pair of human eyes.
  • While the results of just how “awesome” a photo is may not be accurate for every image, the auto-tagging tool could prove useful, generating a list of keywords from object recognition as well as less concrete terms, like love, happiness, and teamwork.
  • Clicking on a keyword will bring up an Everypixel search for other images with that same tag, or users can copy and paste the list of keywords.

Can a computer judge art? A new neural network program will rank photos by their probability of being awesome.
Continue reading “AI can now predict whether or not humans will think your photo is awesome”

Try The Everypixel Tool To See What A Computer Thinks Of Your Best Shot

#AI can now predict whether or not humans will think your photo is awesome

  • The Aesthetics tool, still in beta testing, allows users to upload a photo and get an auto-generated list of tags, as well as a percentage rate on the “chance that this image is awesome.”
  • According to developers, the neural network was trained to view an image much in the same way a human photo editor would, looking at factors such as color, sharpness, and subject.
  • As early users report, the system seems to be fairly good at recognizing factors like whether or not the image is sharp and if the composition is interesting, but it is certainly far from a pair of human eyes.
  • While the results of just how “awesome” a photo is may not be accurate for every image, the auto-tagging tool could prove useful, generating a list of keywords from object recognition as well as less concrete terms, like love, happiness, and teamwork.
  • Clicking on a keyword will bring up an Everypixel search for other images with that same tag, or users can copy and paste the list of keywords.

Can a computer judge art? A new neural network program will rank photos by their probability of being awesome.
Continue reading “Try The Everypixel Tool To See What A Computer Thinks Of Your Best Shot”

Why Artificial Intelligence Is Less Discriminatory Than People

When will #AI stop being so racist? 

 #fintech @valuewalk

  • So it shouldn’t surprise us that algorithms can learn biases from data that has been generated by humans.
  • Because sentencing systems are based on historical data, and black people have historically been arrested and convicted of more crimes, an algorithm could be designed in order to correct for bias that already exists in the system.
  • University of Chicago researcher Berkeley Dietvorst demonstrates that people will avoid using algorithms that make errors, even in cases where statistical forecasts perform better than human forecasts.
  • In addition to examining bias in algorithms, it is critical to examine the bias that pops up when humans use the algorithm’s results to make decisions.
  • So let’s not wring our hands about discrimination in AI models and forget that the alternative of human decision-making in manual processes is much worse.

Artificial intelligence has become so ubiquitous these days that we barely realize when we’re using it. Sophisticated algorithms
Continue reading “Why Artificial Intelligence Is Less Discriminatory Than People”

This interactive map uses machine learning to arrange visually similar fonts

This interactive map uses machine learning to arrange visually similar fonts

  • Typography enthusiasts likely already know how to identify fonts by name, but it’s always useful to explore visually similar fonts when you feel like changing up your options.
  • Design consultant firm IDEO’s Font Map helps you do exactly that, with an interactive tool that lets you browse through fonts by clicking on them and seeing ones nearby that look similar, or by specifically searching for fonts by name.
  • IDEO software designer Kevin Ho built the map using a machine learning algorithm that can sort fonts by visual characteristics, like weight, serif or san-serif, and cursive or non-cursive.
  • “Designers need an easier way to discover alternative fonts with the same aesthetic — so I decided to see if a machine learning algorithm could sort fonts by visual characteristics, and enabling designers to explore type in a new way,” he wrote in a blog post.
  • Services that compare and suggest visually similar fonts already exist, like Identifont and the blog Typewolf, but IDEO’s tool makes it easy to quickly browse and at the very least, appreciate all the options out there that help make the web more beautiful.

Typography enthusiasts likely already know how to identify fonts by name, but it’s always useful to explore visually similar fonts when you feel like changing up your options. Design consultant…
Continue reading “This interactive map uses machine learning to arrange visually similar fonts”