A novel approximate Bayesian filter based on backward stochastic differential equations is introduced. It uses a nonlinear Feynman--Kac representation of the filtering problem and the approximation of an unnormalized filtering density using the well-known deep BSDE method and neural networks. The method is trained offline, which means that it can be applied online with new observations. A mixed a priori-a posteriori error bound is proved under an elliptic condition. The theoretical convergence rate is confirmed in two numerical examples.

For now entrepreneurs are focussing their efforts on humanoid robots for warehouses and factories. The highest profile of those is Elon Musk. His car company, Tesla, is developing a humanoid robot called Optimus. In January he said that "several thousand" will be built this year and he expects them to be doing "useful things" in Tesla factories. Other carmakers are following a similar path.

A numerical scheme for approximating the nonlinear filtering density is introduced and its convergence rate is established, theoretically under a parabolic H\"{o}rmander condition, and empirically for two examples. For the prediction step, between the noisy and partial measurements at discrete times, the scheme approximates the Fokker--Planck equation with a deep splitting scheme, and performs an exact update through Bayes' formula. This results in a classical prediction-update filtering algorithm that operates online for new observation sequences post-training. The algorithm employs a sampling-based Feynman--Kac approach, designed to mitigate the curse of dimensionality. Our convergence proof relies on the Malliavin integration-by-parts formula. As a corollary we obtain the convergence rate for the approximation of the Fokker--Planck equation alone, disconnected from the filtering problem.

Elvis Presley's immersive concert experience is set to leave London all shook up, with an AI rendering of the king of rock'n'roll ready to enthral fans from November 2024. But this is not the first holographic performance – nor will it be the last. Here are some of the other artists whom technology has allowed to tour from beyond the grave, or as their younger selves. Abba's concert kicks off with a lithe and fresh-faced Benny Andersson reassuring the crowd: "This is really me. I just look very good for my age."

And as has become customary in the last few years, major tech companies are taking this week as a chance to share their latest accessibility-minded products. From Apple and Google to Webex and Adobe, the industry's biggest players have launched new features to make their products easier to use. The company actually had a huge set of updates to share, which makes sense since it typically releases most of its accessibility-centric news at this time each year. For 2023, Apple is introducing Assistive Access, which is an accessibility setting that, when turned on, changes the home screen for iPhone and iPad to a layout with fewer distractions and icons. You can choose from a row-based or grid-based layout, and the latter would result in a 2x3 arrangement of large icons.

The purpose of this paper is to explore the use of deep learning for the solution of the nonlinear filtering problem. This is achieved by solving the Zakai equation by a deep splitting method, previously developed for approximate solution of (stochastic) partial differential equations. This is combined with an energy-based model for the approximation of functions by a deep neural network. This results in a computationally fast filter that takes observations as input and that does not require re-training when new observations are received. The method is tested on four examples, two linear in one and twenty dimensions and two nonlinear in one dimension. The method shows promising performance when benchmarked against the Kalman filter and the bootstrap particle filter.

In the next week or so, the sea ice floating atop the Arctic Ocean will shrink to its smallest size this year, as summer-warmed waters eat away at the ice's submerged edges. Record lows for sea ice levels will probably not be broken this year, scientists say. In 2020, the ice covered 3.74 million square kilometers of the Arctic at its lowest point, coming nail-bitingly close to an all-time record low. Currently, sea ice is present in just under 5 million square kilometers of Arctic waters, putting it on track to become the 10th-lowest extent of sea ice in the area since satellite record keeping began in 1979. It's an unexpected finish considering that in early summer, sea ice hit a record low for that time of year. The surprise comes in part because the best current statistical- and physics-based forecasting tools can closely predict sea ice extent only a few weeks in advance, but the accuracy of long-range forecasts falters.

A new artificial intelligence (AI) tool has been developed to enable more accurate prediction of Arctic sea ice conditions months into the future. Led by the British Antarctic Survey (BAS) and the Alan Turing Institute, the research team believes its improved forecasts could underpin new systems to protect Arctic wildlife and coastal communities from the impacts of sea ice loss. The paper has been published in Nature Communications. Sea ice that appears at the North and South poles is difficult to forecast due to its complex relationship with the atmosphere and the ocean below it. The summer Arctic sea ice area has halved over the past four decades due to its sensitivity to increasing temperatures caused by global warming.

Promising more accurate predictions in an era of rapid climate change, a new tool is harnessing deep learning to help better forecast Arctic sea ice conditions months into the future. As described in a paper published in the science journal Nature Communications Thursday, the new AI tool, dubbed IceNet, could lead to improved early-warning systems to protect Arctic wildlife and coastal communities. Created by an international team of researchers led by the British Antarctic Survey and the Alan Turing Institute, IceNet tackles a challenge that has long vexed scientists. "The Arctic is a region on the frontline of climate change and has seen substantial warming over the last 40 years," explained lead author Tom Andersson, a data scientist at the BAS AI Lab, in a statement. "IceNet has the potential to fill an urgent gap in forecasting sea ice for Arctic sustainability efforts and runs thousands of times faster than traditional methods," he added.

Described in the journal Nature Communications, the AI system, IceNet, addresses the challenge of producing accurate Arctic sea ice forecasts for the season ahead – something that has eluded scientists for decades. Sea ice, a vast layer of frozen sea water that appears at the North and South poles, is notoriously difficult to forecast because of its complex relationship with the atmosphere above and ocean below, the researchers said. The sensitivity of sea ice to increasing temperatures has caused the summer Arctic sea ice area to halve over the past four decades, equivalent to the loss of an area around 25 times the size of Great Britain, they said. These accelerating changes, the researchers noted, have dramatic consequences for the world climate, for Arctic ecosystems, and Indigenous and local communities whose livelihoods are tied to the seasonal sea ice cycle. IceNet is almost 95 per cent accurate in predicting whether sea ice will be present two months ahead – better than the leading physics-based model, according to the researchers.