Various approaches have been proposed to learn visuo-motor policies for real-world robotic applications. One solution is first learning in simulation then transferring to the real world. In the transfer, most existing approaches need real-world images with labels. However, the labelling process is often expensive or even impractical in many robotic applications. In this paper, we propose an adversarial discriminative sim-to-real transfer approach to reduce the cost of labelling real data. The effectiveness of the approach is demonstrated with modular networks in a table-top object reaching task where a 7 DoF arm is controlled in velocity mode to reach a blue cuboid in clutter through visual observations. The adversarial transfer approach reduced the labelled real data requirement by 50%. Policies can be transferred to real environments with only 93 labelled and 186 unlabelled real images. The transferred visuo-motor policies are robust to novel (not seen in training) objects in clutter and even a moving target, achieving a 97.8% success rate and 1.8 cm control accuracy.

ANYbotics led the way in the ICRA 2018 Robot Launch Startup Competition on May 22, 2018 at the Brisbane Conference Center in Australia. Although ANYbotics pitched last out of the 10 startups presenting, they clearly won over the judges and audience. As competition winners, ANYbotics received a $3,000 prize from QUT bluebox, Australia's robotics accelerator (currently taking applications for 2018!), plus Silicon Valley Robotics membership and mentoring from The Robotics Hub. ANYbotics is a Swiss startup creating fabulous four legged robots like ANYmal and the core component, the ANYdrive highly integrated modular robotic joint actuator. Founded in 2016 by a group of ETH Zurich engineers, ANYbotics is a spin-off company of the Robotic Systems Lab (RSL), ETH Zurich.

For the first time, new research suggests artificial intelligence may be better than highly-trained humans at detecting skin cancer. A study conducted by an international team of researchers pitted experienced dermatologists against a machine learning system, known as a deep learning convolutional neural network, or CNN, to see which was more effective at detecting malignant melanomas. The results? "Most dermatologists were outperformed by the CNN," the researchers wrote in their report, published in the journal Annals of Oncology. Fifty-eight dermatologists from 17 countries around the world participated in the study. More than half of the doctors were considered expert level with more than five years' experience.

JOURNALISTS at The Boston Globe searched for patterns in public records to uncover priests in the Catholic church who had sexually abused children. Now, researchers think artificial intelligence could do the same job faster, more accurately and on a much wider scale. The Boston Globe investigation, depicted in the film Spotlight, involved looking for clues like priests suddenly going on sick leave or moving around a lot. To continue reading this premium article, subscribe for unlimited access. Existing subscribers, please log in with your email address to link your account access.

Will AI become so intelligent that we don't need government any more? That's the suggestion from Taiwan's digital chief, who notes that "We can replace the entire government by AI in 500 years". That seems unlikely to affect many GovInsider readers – unless modern medicine dramatically increases the average human life expectancy. But already we are seeing this technology improve public service delivery. And far from making government become colder and more mechanised, this ancient institution is getting closer to citizens than ever before.

Human intelligence may have first evolved to help us work together, according to a new study. Research suggests that our ape-like ancestors boosted their brain size when they began to co-operate to hunt for food and shelter. Scientists said that the expanding intelligence of our ancestors in turn helped them better co-operate and take down larger prey, such as mammoths, that they could share with a bigger group. Human intelligence may have first evolved to help us work together, according to a new study. Research suggests that our ape-like ancestors boosted their brain size when they began to co-operate to hunt for food and shelter.

A computer was better than human dermatologists at detecting skin cancer in a study that pitted people against machines in the quest for better, faster diagnostics, researchers said on Tuesday. A team from Germany, the United States and France taught an artificial intelligence system to distinguish dangerous skin lesions from benign ones, showing it more than 100,000 images. The machine – a deep learning convolutional neural network or CNN – was then tested against 58 dermatologists from 17 countries, shown photos of malignant melanomas and benign moles. Just over half the dermatologists were at "expert" level with more than five years of experience, 19% had between two and five years' experience, and 29% were beginners with less than two years under their belt. "Most dermatologists were outperformed by the CNN," the research team wrote in a paper published in the journal Annals of Oncology.

PARIS - A computer was better than human dermatologists at detecting skin cancer in a study that pitted human against machine in the quest for better, faster diagnostics, researchers said Tuesday. A team from Germany, the United States and France taught an artificial intelligence system to distinguish dangerous skin lesions from benign ones, showing it more than 100,000 images. The machine -- a deep learning convolutional neural network or CNN -- was then tested against 58 dermatologists from 17 countries, shown photos of malignant melanomas and benign moles. Just over half the dermatologists were at'expert' level with more than five years of experience, 19 percent had between two and five years' experience, and 29 percent were beginners with less than two years under their belt. 'Most dermatologists were outperformed by the CNN,' the research team wrote in a paper published in the journal Annals of Oncology.

Australian researchers have praised a computer that has been hailed for being better than an international team of specialists at detecting skin cancer. Scientists from Germany, the United States and France developed an artificial intelligence system to distinguish dangerous skin lesions from benign ones, showing it more than 100,000 images. The computer was found to offer more accuracy and fast diagnostics than 58 dermatologists from 17 countries, when shown photos of malignant melanomas and benign moles. Scientists from Germany, the U.S. and France developed an artificial intelligence system to distinguish dangerous skin lesions from benign ones, showing it more than 100,000 images On average, flesh and blood dermatologists accurately detected 86.6 percent of skin cancers from the images, compared to 95 percent for the machine, known as a convolutional neural network or CNN. Australian experts Victoria Mar, from Melbourne's Monash University, and Peter Soyer from the University of Queensland said it was a major breakthrough in detecting skin cancers.

The adoption of machine learning in high-stakes applicatio ns such as healthcare and law has lagged in part because predictions are not accomp anied by explanations comprehensible to the domain user, who often holds ult imate responsibility for decisions and outcomes. In this paper, we propose an appr oach to generate such explanations in which training data is augmented to inc lude, in addition to features and labels, explanations elicited from domain use rs. A joint model is then learned to produce both labels and explanations from the inp ut features. This simple idea ensures that explanations are tailored to the compl exity expectations and domain knowledge of the consumer. Evaluation spans multipl e modeling techniques on a simple game dataset, an image dataset, and a chemi cal odor dataset, showing that our approach is generalizable across domains a nd algorithms. Results demonstrate that meaningful explanations can be reli ably taught to machine learning algorithms, and in some cases, improve modeling ac curacy.