Deep neural networks are susceptible to adversarial attacks. In computer vision,well-crafted perturbations to images can cause neural networks to make mistakes such as confusing a cat with a computer. Previous adversarial attacks have been designed to degrade performance of models or cause machine learning models to produce specific outputs chosen ahead of time by the attacker. We introduce attacks that instead reprogram the target model to perform a task chosen by the attacker—without the attacker needing to specify or compute the desired output for each test-time input. This attack finds a single adversarial perturbation, that can be added to all test-time inputs to a machine learning model in order to cause the model to perform a task chosen by the adversary—even if the model was not trained to do this task. These perturbations can thus be considered a program for the new task. We demonstrate adversarial reprogramming on six ImageNet classification models, repurposing these models to perform a counting task, as well as classification tasks: classification of MNIST and CIFAR-10 examples presented as inputs to the ImageNet model. Read More
Monthly Archives: July 2019
The “one program” hypothesis
The Future of Robotics and Artificial Intelligence (Andrew Ng, Stanford University, STAN 2011)
One program hypothesis discussion starts around the 8:30 mark.
Cyber Security Threats
AI ‘emotion recognition’ can’t be trusted
As artificial intelligence is used to make more decisions about our lives, engineers have sought out ways to make it more emotionally intelligent. That means automating some of the emotional tasks that come naturally to humans — most notably, looking at a person’s face and knowing how they feel.
To achieve this, tech companies like Microsoft, IBM, and Amazon all sell what they call “emotion recognition” algorithms, which infer how people feel based on facial analysis. For example, if someone has a furrowed brow and pursed lips, it means they’re angry. If their eyes are wide, their eyebrows are raised, and their mouth is stretched, it means they’re afraid, and so on.
But the belief that we can easily infer how people feel based on how they look is controversial, and a significant new review of the research suggests there’s no firm scientific justification for it. Read More
Is Your Data Ready for AI?
Companies are champing at the bit to introduce any solution that promises Artificial Intelligence and Machine Learning. But hasty adoption is leaving one important question unanswered.
Is your data ready for AI?
For most companies, the answer is no.
Read More
Measuring Progress Toward AGI Is Hard
Artificial General Intelligence (AGI) is still a ways off in the future but surprisingly there’s been very little conversation about how to measure if we’re getting close. This article reviews a proposal to benchmark existing AIs against animal capabilities in an Animal-AI Olympics. It’s a real thing and just now accepting entrants. Read More
Vacation scammer speaks with a Jolly Roger bot named Ox-Gut McGee
5 Simple Full Stack Data Science Projects To Put On Your Resume
Whether large or small, almost every organisation is looking for aspiring data scientists who will not only help them churn out meaningful insights from data but also help them stay ahead of the curve.
It does not matter if you are a college drop-out or a fresher, with the right knowledge of tools and a good understanding of the concepts of machine learning you can still pursue a fruitful data science career with a good pay scale. Read More
AI And Chatbots Are Transforming The Customer Experience
Artificial Intelligence (AI) is dramatically changing business, and chatbots, fueled by AI, are becoming a viable customer service channel. The best ones deliver a customer experience (CX) in which customers cannot tell if they are communicating with a human or a computer. AI has come a long way in recognizing the content – and context – of customers’ requests and questions. Read More
Gallery Go: a fast, helpful way to organize your photos offline
Today, at Google for Nigeria we introduced Gallery Go: a photo gallery, designed to work offline, that uses machine learning to automatically organize and make your photos look their best. Gallery Go helps first time smartphone owners easily find, edit, and manage photos, without the need for access to high-speed internet or cloud backup.
Gallery Go automatically organizes your photos by the people and things you take photos of, so you can easily find your favorite selfie, remember where you had the best puff puff, and keep track of important documents. You don’t have to manually label your photos and all these features run on your phone, without using your data. You can create folders to organize your photos, and Gallery Go works with SD cards, so you can easily copy them from your phone. Read More
Rick's Cafe AI 