Stocks lost further ground on the Tokyo Stock Exchange on Tuesday, weighed down by the yen's ascent. The 225-issue Nikkei average fell 65.71 points, or 0.34 percent, to close at 19,455.88. On Friday, the key market gauge gave up 68.55 points. The Tokyo market was closed on Monday for a national holiday. The Topix index of all first-section issues ended down 2.43 points, or 0.16 percent, at 1,563.42, after shedding 6.84 points on Friday.

The issue of sustainability has never been more prominent. Around the world, headlines are full of warnings on the dangers of climate change as companies, people and governments campaign for greener policies and practices. One of the sectors most affected by this drive is chemical processing and manufacturing. Every year, more than a thousand new chemical substances are introduced into the U.S. For each one, the potential applications need to be weighed against myriad potential health and environmental impacts across a broad range of metrics, such as energy consumption, toxicity or biodegradability across product lifecycles. If chemical processors and manufacturers are able to shift toward more sustainable practices – e.g., processes that are more energy efficient, require lower input volumes and are more environmentally and biologically-friendly – the benefits for both the industry and the environment would be significant.

It is generally accepted that all models are wrong -- the difficulty is determining which are useful. Here, a useful model is considered as one that is capable of combining data and expert knowledge, through an inversion or calibration process, to adequately characterize the uncertainty in predictions of interest. This paper derives conditions that specify which simplified models are useful and how they should be calibrated. To start, the notion of an optimal simplification is defined. This relates the model simplifications to the nature of the data and predictions, and determines when a standard probabilistic calibration scheme is capable of accurately characterizing uncertainty. Furthermore, two additional conditions are defined for suboptimal models that determine when the simplifications can be safely ignored. The first allows a suboptimally simplified model to be used in a way that replicates the performance of an optimal model. This is achieved through the judicial selection of a prior term for the calibration process that explicitly includes the nature of the data, predictions and modelling simplifications. The second considers the dependency structure between the predictions and the available data to gain insights into when the simplifications can be overcome by using the right calibration data. Furthermore, the derived conditions are related to the commonly used calibration schemes based on Tikhonov and subspace regularization. To allow concrete insights to be obtained, the analysis is performed under a linear expansion of the model equations and where the predictive uncertainty is characterized via second order moments only.

Information in neural networks is represented as weighted connections, or synapses, between neurons. This poses a problem as the primary computational bottleneck for neural networks is the vector-matrix multiply when inputs are multiplied by the neural network weights. Conventional processing architectures are not well suited for simulating neural networks, often requiring large amounts of energy and time. Additionally, synapses in biological neural networks are not binary connections, but exhibit a nonlinear response function as neurotransmitters are emitted and diffuse between neurons. Inspired by neuroscience principles, we present a digital neuromorphic architecture, the Spiking Temporal Processing Unit (STPU), capable of modeling arbitrary complex synaptic response functions without requiring additional hardware components. We consider the paradigm of spiking neurons with temporally coded information as opposed to non-spiking rate coded neurons used in most neural networks. In this paradigm we examine liquid state machines applied to speech recognition and show how a liquid state machine with temporal dynamics maps onto the STPU-demonstrating the flexibility and efficiency of the STPU for instantiating neural algorithms.

The key to trading in 2017 is that once you make money, keep it. Exercise iron risk control, and don't take marginal trades. You are trying to make yourself rich, not your broker. Don't do stupid things either, like selling short naked deep out-of-the-money put options which I know 99% of the newsletters out there recommend for easy money. The people who first subscribed to the Mad Hedge Fund Trader a year ago certainly are happy.

Declining birth rate, aging population, natural disasters, pollution: Do these sound like issues the IT industry can deal with? Japanese businesses say yes, and a number of them are at the Cebit trade show in Hanover, Germany, to explain why. Industry 4.0 -- the building of "smart factories" in which machines monitor one another and make decentralized decisions about production and maintenance -- has been a theme of recent Cebit shows. Now, under the banner Society 5.0, the show's partner country for 2017, Japan, wants to take the transformation beyond industry, making "smart society" one of the show's talking points. Behind the drive are some very real societal problems.

Model predictive control (MPC) is a popular control method that has proved effective for robotics, among other fields. MPC performs re-planning at every time step. Re-planning is done with a limited horizon per computational and real-time constraints and often also for robustness to potential model errors. However, the limited horizon leads to suboptimal performance. In this work, we consider the iterative learning setting, where the same task can be repeated several times, and propose a policy improvement scheme for MPC. The main idea is that between executions we can, offline, run MPC with a longer horizon, resulting in a hindsight plan. To bring the next real-world execution closer to the hindsight plan, our approach learns to re-shape the original cost function with the goal of satisfying the following property: short horizon planning (as realistic during real executions) with respect to the shaped cost should result in mimicking the hindsight plan. This effectively consolidates long-term reasoning into the short-horizon planning. We empirically evaluate our approach in contact-rich manipulation tasks both in simulated and real environments, such as peg insertion by a real PR2 robot.

The technologies making waves in 2017 include brain implants and quantum computers. Here is a list of the top 10 technologies that are expected to be prevalent this year, according to MIT. At the top of the list is behavior-reinforced artificial intelligence. Whether that's mastering the complex game of Go and beating a champion or learning to merge a self-driving car into traffic. The technology is based on reinforcement learning, documented more than a 100 years ago by psychologist Edward Thorndike.

Samsung's'make or break' Galaxy S8 smartphone will have a smart assistant called Bixby to take on Apple's Siri, the firm has accidentally revealed on its own website. An Italian page found on the Samsung website clearly states'comandi vocali con Bixby', which means'voice commands with Bixby' when translated into English. Bixby is believed to be central to Samsung's plan for the S8, with the handset even having a dedicated button for the assistant. Samsung's Galaxy S7: The latest version of its flagship phone will come in two sizes, both with curved screens. An Italian page found on the Samsung website clearly states'comandi vocali con Bixby', which means'voice commands with Bixby' when translated into English. Bixby could be used for a wide variety of functions in a similar way to Apple's Siri.

The much-heralded technological singularity will happen in 2029, according to Google's director of engineering. Ray Kurzweil, a futurist who has made a name for himself through his predictions, shared his thoughts about what's in store for humans and machines in an interview with SXSW in Texas. He believes that the so-called singularity – the moment when artificial intelligence exceeds man's intellectual capacity and creates a runaway effect, which many believe will lead to the demise of the human race – is little over a decade away. "By 2029, computers will have human-level intelligence," said Mr Kurzweil. "That leads to computers having human intelligence, our putting them inside our brains, connecting them to the cloud, expanding who we are. It's here, in part, and it's going to accelerate."