Tyler O'Neal, Staff Editor CHEMISTRY

Using MDP, UCL Mathematics prof builds AI that predicts when a bank should be bailed out

An artificial intelligence tool developed by researchers at UCL and the Queen Mary University of London could help governments decide whether or not to bail out a bank in crisis by predicting if the intervention will save money for taxpayers in the long term.

The AI tool assesses not only if a bailout is the best strategy for taxpayers, but also suggests how much should be invested in the bank, and which bank or banks should be bailed out at any given time.

The algorithm was tested by the authors using data from the European Banking Authority on a network of 35 European financial institutions judged to be the most important to the global financial system, but it can also be used and calibrated by national banks using detailed proprietary data unavailable to the public.

Dr. Neofytos Rodosthenous (UCL Mathematics), the corresponding author of the paper, said: “Government bank bailouts are complex decisions that have financial, social, and political implications. We believe the AI approach we have developed can be an important tool for governments, helping officials assess specifically financial implications – this means checking if a bailout is in the best interest of taxpayers, or whether it would be better value for money to let the bank fail. Our techniques are freely available for banking authorities to use as tools in their decision-making process.”

Co-author Professor Vito Latora (Queen Mary University of London) added: “Governments and banking authorities can also use our approach to retrospectively review past crises and gain valuable learnings to inform future actions. One could, for example, review the UK government bailout of the Royal Bank of Scotland (RBS) during the financial crisis of 2007-9 and reflect on how this could potentially be improved (from a financial standpoint) in the future in order to primarily benefit taxpayers.”

In a bank bailout, government investment in a bank increases the bank’s equity and reduces its risk of default. This cost in the short term may be justified to the taxpayer if it leads to lower taxpayer losses in the long term – i.e., it prevents bank defaults that are more damaging to government finances.

In their study, the researchers created a mathematical framework for comparing different bailout strategies in terms of predicted losses to taxpayers. Considered factors include how long the financial crisis is expected to last, the likelihood of each bank defaulting and the effect of a default on other banks in the network, as well as taxpayers’ stakes in the banks.

Using a mathematical control process, called Markov Decision Process (MDP), the researchers incorporated into this framework the effect of government intervention at any given point in time.

They then developed a bespoke AI algorithm to assess optimal bailout strategies, comparing no intervention to different types of intervention – that is, varying levels of investment in one bank or many banks – at different time points during a crisis. An AI technique is needed as modeling such a system is highly complex, as the future behavior of all banks in the system can be infinite.

In their case study using data from the European Banking Authority, they showed that a government bailout would be optimal only if the taxpayers’ stakes in the banks were greater than some critical threshold value, determined via the model. The optimal policy drastically changed once the percentage loss had gone above this threshold.

Moreover, it was shown that government bailout tended to be more favorable the greater the network’s distress (defined in terms of a percentage reduction in the banks’ equity), the longer the crisis lasted and the bigger the banks’ exposures to other banks were (that is, how much they had lent other banks and therefore stood to lose if these banks failed).

The researchers also found that, once a bank had received a bailout, the best strategy for taxpayers was if the government continued to invest in that bank to prevent default. This could lead to a lack of incentive for the rescued bank to guard against risk, potentially increasing risk-taking.

Lead author Dr. Daniele Petrone said: “Banks have so far weathered the current economic storm triggered by the Covid-19 pandemic. Their resilience has been bolstered by regulatory measures introduced following the global financial crisis of 2007-9 and by accommodating central banks' monetary policies that have avoided bankruptcies across industries. However, no one can predict the effect on the financial system as central banks reverse previous policies, such as increasing interest rates due to inflation concerns, and so bailouts are still a possibility.”

Friborg shows how Arctic vegetation has a major impact on warming

Tyler O'Neal, Staff Editor CHEMISTRY

An international team of research scientists, University of Copenhagen researchers among them, has documented the central role of vegetation in Arctic warming for the first time. The new results allow us to make more precise climate predictions, with the researchers pointing out that current models remain flawed.

The Arctic is warming three times faster than the global average. In areas where snow and ice used to reflect sunlight back into the atmosphere, melted terrain now absorbs heat into the earth’s surface. There has long been speculation about the degree to which vegetation emerging from melting snow has on warming, compared to other factors like snow, precipitation, cloud cover, and geographical location.

Numerous studies have already demonstrated the significance of greenhouse gas emissions, such as from CO2 and methane, in Arctic ecosystems. Far fewer have investigated the influence of vegetation on Arctic climates. Using new analyses of data measured at 64 Arctic sites from 1994-2021, an international research team has become the first to document the great importance of vegetation for Arctic warming. 

“Theoretically, it has long been understood that surface vegetation helps heat an area as plants absorb solar radiation. In our new study, we confirm this theory through actual measurements and demonstrate a correlation between the amount of energy absorbed on the surface and the types of vegetation found there,” says Professor Thomas Friborg of the Department of Geosciences and Natural Resource Management.

For more than 20 years, Friborg has measured climate data from northern Sweden, northern Russia, and Greenland, among other places, and has contributed data to the study.  

Theory confirmed

Among other things, the researchers compared fifteen factors that affect the so-called “surface energy budget” (SEB), which describes how solar energy is converted when it hits the earth's surface. In doing so, the researchers studied how various Arctic areas such as barren tundra, peat bogs, shrub-covered tundra, and wetlands influence how solar energy is converted.

The results demonstrate that some of the greatest differences in energy conversion are found between dry areas with little vegetation, where grasses and lichens typically grow, and wet areas such as peat bogs, that are rich in mosses, shrubs, and small trees. Dry soil surfaces produce greater warming than wet areas as the energy from wet areas is converted into evaporation.

This is just one example of the various roles that vegetation types play in the warming of an area, differences that today’s climate models still fail to take fully into account.

"Our study shows that the type of vegetation on an Arctic surface has a major impact on how to direct warming will be. Whether there are shrubs, grasses, mosses or wetlands matters considerably for the degree to which solar energy is absorbed and how it gets converted. In fact, in some cases, the vegetation type is nearly as decisive as whether snow is present," says Professor Friborg.

Crucial for climate predictions

Several studies have shown that the Arctic is greening as temperatures rise due to what is known as arctic amplification. As this leads to the emergence of more vegetation, knowing how vegetation reacts with sunlight and affects warming is crucial for climate predictions. It is here that the researchers contributed new and important knowledge.

"It is a very large study, and an important observation of the way Arctic plants convert solar energy. The results are very likely to influence our way of predicting climatic changes in the Arctic and globally because we can now put a few values on vegetation-related differences," says Friborg.

The researchers’ findings also highlight the potential for improving our current predictions for how climate evolves. Current knowledge remains checkered, and ever more data collection is needed to understand this intricate puzzle. 

“In many ways, the Arctic is the canary in the coal mine – it is where we first and most powerfully see global warming. But at the same time, it is incredibly complex to predict. We are currently witnessing warming of 3-4 degrees, which is higher than quite a few of the models predicted 20 years ago. As such, there is a constant need to refine models and include as much data as possible in them,” concludes Professor Thomas Friborg.

The Hebrew University of Jerusalem utilizes supercomputerized simulations of the atmospheres of exoplanets to identify those suitable for human settlements

Tyler O'Neal, Staff Editor CHEMISTRY

Climatology of (a, c, e) maximum temperature (Tmax in °C) and (b, d, f) total precipitation (Prec in mm d−1) for TRAPPIST-1e. The climatology is computed using the mean of 80 yr ExoCAM simulations with varying pCO2. The different pCO2 scenarios are described in Section 3.1 and are referred to as “Low” (10−2 Bar), “Mid” (10−1 Bar), and “High” (1 Bar). In panel (a) the subregions used in Figures 7–10 and 12 are marked with gray rectangles. The subregions considered are global (−90–90N, 0–360E), mid-latitude antistellar (30-60N, 345-15E), mid-latitude substellar (30-60N, 165-195E), equatorial antistellar (−15-15N, 345-15E), equatorial substellar (−15-15N, 165-195E), equatorial west-terminator (−15-15N, 75-105E), and equatorial east-terminator (−15-15N, 255-285E).The climate crisis presents a huge challenge to all people on Earth. It has led many scientists to look for exoplanets, planets outside our solar system that humans could potentially settle. The James Webb Space Telescope was developed as part of this search to provide detailed observational data about earth-like exo-planets in the coming years. A new project, led by Dr. Assaf Hochman at the Fredy & Nadine Herrmann Institute of Earth Sciences at the Hebrew University of Jerusalem (HU) has successfully developed a framework to study the atmospheres of distant planets and locate those planets fit for human habitation, without having to visit them physically. Their research study was published in the academic journal Astrophysical Journal.

Classifying climate conditions and measuring climate sensitivity are central elements when assessing the viability of exoplanets as potential candidates for human habitation. In the current study, the research team examined TRAPPIST-1e, a planet located some 40 light years from the Earth and scheduled to be documented by the James Webb Space Telescope in the coming year. The researchers looked at the sensitivity of the planet’s climate to increases in greenhouse gases and compared it with conditions on Earth. Using a supercomputerized simulation of the climate on TRAPPIST-1e, they could assess the impact of changes in greenhouse gas concentration.

The study focused on the effect of an increase in carbon dioxide on extreme weather conditions, and on the rate of changes in weather on the planet. “These two variables are crucial for the existence of life on other planets, and they are now being studied in depth for the first time in history,” explained Hochman.

According to the research team, studying the climate variability of earth-like exo-planets provides a better understanding of the climate changes we are currently experiencing on Earth. Additionally, this kind of research offers a new understanding of how planet Earth’s atmosphere might change in the future.

Hochman and his research partners found that planet TRAPPIST-1e has a significantly more sensitive atmosphere than planet Earth. They estimate that an increase in greenhouse gases there could lead to more extreme climate changes than we would experience here on Earth because one side of TRAPPIST-1e constantly faces its own sun, in the same way, that our moon always has one side facing the Earth.

As Hochman concluded, “the research framework we developed, along with observational data from the Webb Space Telescope, will enable scientists to efficiently assess the atmospheres of many other planets without having to send a space crew to visit them physically. This will help us make informed decisions in the future about which planets are good candidates for human settlement and perhaps even to find life on those planets.”