Australia's first driverless shuttle trial turned a year old this week, during which the bus has clocked up more than 4200 km in autonomous mode. The RAC Intellibus has made around 1500 thirty-minute trips, carrying more than 4300 passengers around its route on open road in South Perth. The autonomous bus – which can reach speeds of 45km per hour, but averages at around 25km per hour – is fully electric and uses light detection and ranging (LIDAR), stereovision cameras, GPS, odometry and autonomous emergency braking to detect and avoid obstacles and maintain its course. It is considered to have'Level 4' automation (as defined by SAE International standards) which means the vehicle can perform all safety critical driving functions without any occupants. Nevertheless, the bus has a'chaperone' whom can take the wheel (actually a Playstation controller) if needed.

Since the days of Da Vinci's "Ornithoper", mankind's greatest minds have sought inspiration from the natural world for their technological creations. It's no different in the modern world, where bleeding-edge advancements in machine learning and artificial intelligence have begun taking their design cues from the most advanced computational organ in the natural word: the human brain. Deep learning neural networks -- the likes of which power AlphaGo as well as the current generation of image recognition and language translation systems -- are the best machine learning systems we've developed to date. They're capable of incredible feats but still face significant technological hurdles, like the fact that in order to be trained on a specific skill they require upfront access to massive data sets. What's more if you want to retrain that neural network to perform a new skill, you've essentially got to wipe its memory and start over from scratch -- a process known as "catastrophic forgetting".

Since the days of Da Vinci's "Ornithoper", mankind's greatest minds have sought inspiration from the natural world for their technological creations. It's no different in the modern world, where bleeding-edge advancements in machine learning and artificial intelligence have begun taking their design cues from the most advanced computational organ in the natural word: the human brain. Deep learning neural networks -- the likes of which power AlphaGo as well as the current generation of image recognition and language translation systems -- are the best machine learning systems we've developed to date. They're capable of incredible feats but still face significant technological hurdles, like the fact that in order to be trained on a specific skill they require upfront access to massive data sets. What's more if you want to retrain that neural network to perform a new skill, you've essentially got to wipe its memory and start over from scratch -- a process known as "catastrophic forgetting". Compare that to the human brain, which learns incrementally rather than bursting forth fully-formed from a sea of data points.

Drones are harnessing artificial intelligence to detect sharks approaching Australian beaches. Starting next month, Little Ripper drones will be able to monitor sharks in real-time with approximately 90 percent accuracy. By contrast, humans are only about 20 to 30 percent accurate when spotting sharks. "It's not about replacing human beings all together, it's about assisting human beings to get the work done in a better way with more accuracy," Dr. Nabin Sharma, a research associate at the University of Technology Sydney's School of Software said in an interview with Reuters. "That's what the application is meant for."

It will not be the first experiment with advanced pizza delivery technology. Australia-based Domino's Pizza Enterprises, the Ann Arbor-based company's largest independent franchisee, has tested delivery to customers in New Zealand via drone and self-driving robot.

Ford Motor Co and Domino's Pizza Inc in September will begin testing Michigan consumers' reactions to having their pies delivered by self-driving vehicles, the companies said on Tuesday. It will not be the first experiment with advanced pizza delivery technology. Australia-based Domino's Pizza Enterprises, the Ann Arbor-based company's largest independent franchisee, has tested delivery to customers in New Zealand via drone and self-driving robot. In a blog post last week, Sherif Marakby, head of Ford's autonomous and electric vehicles, signaled the automaker's broader ambitions, saying Ford planned to cooperate "with multiple partners" in deploying self-driving vehicles "designed to improve the movement of people and goods." Previously, Ford executives had said the company expected to launch a self-driving shuttle for commercial ride-sharing fleets in 2021.

IBM Watson Internet of Things 1,760 views IBM Watson presents Soul Machines, LENDIT Conference 2017 (Professional Camera) - Duration: 5:12.

This paper presents a constructive algorithm that achieves successful one-shot learning of hidden spike-patterns in a competitive detection task. It has previously been shown (Masquelier et al., 2008) that spike-timing-dependent plasticity (STDP) and lateral inhibition can result in neurons competitively tuned to repeating spike-patterns concealed in high rates of overall presynaptic activity. One-shot construction of neurons with synapse weights calculated as estimates of converged STDP outcomes results in immediate selective detection of hidden spike-patterns. The capability of continual learning is demonstrated through the successful one-shot detection of new sets of spike-patterns introduced after long intervals in the simulation time. Simulation expansion (Lightheart et al., 2013) has been proposed as an approach to the development of constructive algorithms that are compatible with simulations of biological neural networks. A simulation of a biological neural network may have orders of magnitude fewer neurons and connections than the related biological neural systems; therefore, simulated neural networks can be assumed to be a subset of a larger neural system. The constructive algorithm is developed using simulation expansion concepts to perform an operation equivalent to the exchange of neurons between the simulation and the larger hypothetical neural system. The dynamic selection of neurons to simulate within a larger neural system (hypothetical or stored in memory) may be a starting point for a wide range of developments and applications in machine learning and the simulation of biology.

The generation of preferences represented as CP-nets for experiments and empirical testing has typically been done in an ad hoc manner that may have introduced a large statistical bias in previous experimental work. We present novel polynomial-time algorithms for generating CP-nets with n nodes and maximum in-degree c uniformly at random. We extend this result to several statistical cultures commonly used in the social choice and preference reasoning literature. A CP-net is composed of both a graph and underlying cp-statements; our algorithm is the first to provably generate both the graph structure and cp-statements, and hence the underlying preference orders themselves, uniformly at random. We have released this code as a free and open source project. We use the uniform generation algorithm to investigate the maximum and expected flipping lengths, i.e., the maximum length over all outcomes o and o', of a minimal proof that o is preferred to o'. Using our new statistical evidence, we conjecture that, for CP-nets with binary variables and complete conditional preference tables, the expected flipping length is polynomial in the number of preference variables. This has positive implications for the usability of CP-nets as compact preference models.

Starting next month, Australia will deploy drones powered by artificial intelligence to patrol Australian beaches for sharks. As reported by Reuters, publicly available videos are used to train the system's algorithms and differentiate sharks from other marine creatures, surfers, swimmers and boats. "Studies have shown that people have a 20-30 percent accuracy rate when interpreting data from aerial images to detect shark activity. Detection software can boost that rate to 90 percent," said Dr. Nabin Sharma, a research associate at the University of Technology Sydney's School of Software. "It's not about replacing human beings all together, it's about assisting human beings to get the work done in a better way with more accuracy. That's what the application is meant for."