New Delhi/Kolkata, India – A pair of black Dandy Pik Pik loafers covered in sharp, uneven spikes and shiny studs was part of the evidence before Judge Pratibha M Singh in an intellectual-property lawsuit brought by French luxury shoe brand Christian Louboutin against an Indian shoe manufacturer in a Delhi high court last year. Louboutin's lawyers had already regaled the court with anecdotes about the iconic status of their shoes. The signature stilettos, with their luxuriant red soles, had starred in movies like The Devil Wears Prada and Sex and The City, and were registered as a trademark in India and other countries, they said. Riding on the brand's reputation, the lawyers were now trying to make the point that spiked shoes, too, were unique to Christian Louboutin, and the defendant, Shutiq – The Shoe Boutique, was manufacturing and selling their designs in India illegally. Incriminating evidence presented to Judge Singh included testimony from ChatGPT, saying that Christian Louboutin is known for spiked men's shoes. Then there were photographs of Shutiq's 26 spiked and bedazzled shoes next to Louboutin originals, including Dandy Pik Pik.

Let's picture the following game; two individuals, an outstanding counterfeiter who is well known for producing the best fake bills ever made, and the other hand, police, who are responsible for identifying whether the money moving around is real or fake. We can cast Tom and Leo for the movie, right? Now, how is this related to Machine Learning, Deep Learning, or any kind of automated Learning known by humankind? The promise of deep learning is to discover rich, hierarchical models that represent probability distributions over the kinds of data encountered in artificial intelligence applications, such as natural images, audio waveforms containing speech, and symbols, among others. We dreamed about Deep Learning implementations that can actually create rather than copy, and that my friends, is part of this journey.

Generative adversarial networks (GANs), formed in 2014 [1], is a state of the art deep neural network with many applications. Unlike the traditional machine learning in unsupervised learning (it does not require a targeted label), GANs is a generative model which generates new content by given data. The analogy of GANs is known as a fake-currency detection game between a counterfeiter and police [1]. According to the tutorial of GANs by Goodfellow [2], GANs consists of two characters, namely, the generator (counterfeiter) and the discriminator (police). The counterfeiter tries to produce fake money and deceive the police (discriminator) by looking at the real banknote.

Generative Adversarial Networks (GANs) are a revolutionary class of Deep Neural Networks (DNNs) that have been successfully used to generate realistic images, music, text, and other data. However, it is well known that GAN training can be notoriously resource-intensive and presents many challenges. Further, a potential weakness in GANs is that discriminator DNNs typically provide only one value (loss) of corrective feedback to generator DNNs (namely, the discriminator's assessment of the generated example). By contrast, we propose a new class of GAN we refer to as LogicGAN, that leverages recent advances in (logic-based) explainable AI (xAI) systems to provide a "richer" form of corrective feedback from discriminators to generators. Specifically, we modify the gradient descent process using xAI systems that specify the reason as to why the discriminator made the classification it did, thus providing the richer corrective feedback that helps the generator to better fool the discriminator. Using our approach, we show that LogicGANs learn much faster on MNIST data, achieving an improvement in data efficiency of 45% in single and 12.73% in multi-class setting over standard GANs while maintaining the same quality as measured by Fr\'echet Inception Distance. Further, we argue that LogicGAN enables users greater control over how models learn than standard GAN systems.

Counterfeiting continues to be a global concern for brands. Indeed, trade in counterfeit and pirated goods has risen steadily in the last few years -- even as overall trade volumes stagnated -- and now stands at 3.3% of global trade, according to a new report by the OECD and the EU's Intellectual Property Office[1]. Brands keep spending a significant amount of money trying to fight counterfeiters but so far it seems to be a losing fight. For several reasons, brands must react. It has been proven that counterfeit consumer goods do destroy a brand reputation, negatively affect consumer confidence or result in claims and damages against manufacturers.

Based on its core scientist team's top-level R&D strength, as well as technological innovation and breakthroughs, Squirrel AI Learning started holding four "man-machine competitions" in Zhengzhou, Chengdu and Dongying in October 2017 in a bid to identify any difference between its adaptive learning system and human teaching. Dr. Kalns demonstrated to the RE-WORK audience the results of the four competitions: surprisingly, machine teaching outperformed human teaching in all the four competitions. Taking the fourth competition, which unfolded in one hundred cities, as an example, students at the same intellectual level were divided into two groups and received human teaching and Squirrel AI Learning respectively. Every student in the machine teaching group learned 42 knowledge points on the average, while every student in the human teaching learned 28 knowledge points on the average; in terms of average scoring in the core part of the competition, the students in the AI teaching group had their scores increased by 5.4 on the average, while the students in the human teaching group just had their scores increased by 0.7 on the average, suggesting that machine teaching enabled students to take a firmer grasp of knowledge points than human teaching and improved the learning efficiency more significantly than human teaching. According to the results, Squirrel AI Learning is basically the same as or better than individualized human teaching.

Artificial intelligence is being credited for helping Hong Kong customs officials increase seizures of fake goods sold online by about one-third in the first six months of this year, resulting in a haul of counterfeit items worth HK$1.96 million (US$247,000). A new supercomputer they began using last December scoured websites 24 hours a day and detected close to 2,000 of the 5,200 items seized by the Customs & Excise Department. Over the same period last year, officers netted 11,800 pieces of counterfeit goods worth HK$1.47 million. A source said the department might look into expanding the capacity of the computer, which gathers important information during investigations, but he stressed it would complement rather than replace manual enforcement work by customs officers. "The analytics tool saves us a lot of time screening online platforms manually," the source said.

That's the rough amount of money that counterfeiters displaced last year by selling phony products. Some 2.5% of all trade is for fake goods. The United States is hit hardest by the scourge of counterfeit products -- U.S. brands accounted in 2013 for 20% of the world's infringed intellectual property. When most people think about counterfeiting, they think of knock-off Louis Vuitton handbags sold on the sidewalk. But fake products also include business and enterprise products, as well as everyday consumer goods.

Luxury handbags may exude the confidence of wealth, but take them with a grain of fancy salt -- the counterfeit business is booming. Around 2.5 percent of global imports to the US are counterfeit or pirated goods. This amounts to about half a trillion dollars' worth of merchandise, and these ill-gotten gains fund organized crime. According to Maysa Razavi, an attorney with the International Trademark Association, shoppers often don't know that "the same people who are counterfeiting are involved in human trafficking and terrorism." The next time you don a head-to-toe Gucci ensemble, keep in mind that the clutch you bought for a bargain could be lining the pockets of an underworld kingpin.

In 2014, the research paper Generative Adversarial Nets (GAN) by Goodfellow et al. was a breakthrough in the field of generative models. Leading researcher Yann Lecun himself called adversarial nets "the coolest idea in machine learning in the last twenty years." Today, thanks to this architecture, we're going to build an AI that generates realistic pictures of cats. To view the full working code, see my Github repository. It will help if you already have some experience in Python, Deep Learning and Tensorflow, and CNNs (Convolutional Neural Nets).