̽��ֱ�� of Cambridge - ̽��ֱ��Alan Turing Institute

Fully AI driven weather prediction system could start revolution in forecasting

sc604 — Thu, 20 Mar 2025 15:56:54 +0000

̽��ֱ��system, Aardvark Weather, has been supported by the Alan Turing Institute, Microsoft Research and the European Centre for Medium Range Weather Forecasts. It provides a blueprint for a new approach to weather forecasting with the potential to transform current practices. The results are reported in the journal Nature.

“Aardvark reimagines current weather prediction methods offering the potential to make weather forecasts faster, cheaper, more flexible and more accurate than ever before, helping to transform weather prediction in both developed and developing countries,” said Professor Richard Turner from Cambridge’s Department of Engineering, who led the research. “Aardvark is thousands of times faster than all previous weather forecasting methods.”

Current weather forecasts are generated through a complex set of stages, each taking several hours to run on powerful supercomputers. Aside from daily usage, the development, maintenance and use of these systems require significant time and large teams of experts.

More recently, research by Huawei, Google, and Microsoft has shown that one component of the weather forecasting pipeline, the numerical solver (which calculates how weather evolves over time), can be replaced with AI, resulting in faster and more accurate predictions. This combination of AI and traditional approaches is now being used by the European Centre for Medium Range Weather Forecasts (ECMWF).

But with Aardvark, researchers have replaced the entire weather prediction pipeline with a single, simple machine learning model. ̽��ֱ��new model takes in observations from satellites, weather stations and other sensors and outputs both global and local forecasts.

This fully AI driven approach means predictions that were once produced using many models – each requiring a supercomputer and a large support team to run – can now be produced in minutes on a desktop computer.

When using just 10% of the input data of existing systems, Aardvark already outperforms the United States national GFS forecasting system on many variables. It is also competitive with United States Weather Service forecasts that use input from dozens of weather models and analysis by expert human forecasters.

“These results are just the beginning of what Aardvark can achieve,” said first author Anna Allen, from Cambridge’s Department of Computer Science and Technology. “This end-to-end learning approach can be easily applied to other weather forecasting problems, for example hurricanes, wildfires, and tornadoes. Beyond weather, its applications extend to broader Earth system forecasting, including air quality, ocean dynamics, and sea ice prediction.”

̽��ֱ��researchers say that one of the most exciting aspects of Aardvark is its flexibility and simple design. Because it learns directly from data it can be quickly adapted to produce bespoke forecasts for specific industries or locations, whether that's predicting temperatures for African agriculture or wind speeds for a renewable energy company in Europe.

This contrasts to traditional weather prediction systems where creating a customised system takes years of work by large teams of researchers.

“ ̽��ֱ��weather forecasting systems we all rely on have been developed over decades, but in just 18 months, we’ve been able to build something that’s competitive with the best of these systems, using just a tenth of the data on a desktop computer,” said Turner, who is also Lead Researcher for Weather Prediction at the Alan Turing Institute.

This capability has the potential to transform weather prediction in developing countries where access to the expertise and computational resources required to develop conventional systems is not typically available.

“Unleashing AI’s potential will transform decision-making for everyone from policymakers and emergency planners to industries that rely on accurate weather forecasts,” said Dr Scott Hosking from ̽��ֱ��Alan Turing Institute. “Aardvark’s breakthrough is not just about speed, it’s about access. By shifting weather prediction from supercomputers to desktop computers, we can democratise forecasting, making these powerful technologies available to developing nations and data-sparse regions around the world.”

“Aardvark would not have been possible without decades of physical-model development by the community, and we are particularly indebted to ECMWF for their ERA5 dataset which is essential for training Aardvark,” said Turner.

“It is essential that academia and industry work together to address technological challenges and leverage new opportunities that AI offers,” said Matthew Chantry from ECMWF. “Aardvark’s approach combines both modularity with end-to-end forecasting optimisation, ensuring effective use of the available datasets."

“Aardvark represents not only an important achievement in AI weather prediction but it also reflects the power of collaboration and bringing the research community together to improve and apply AI technology in meaningful ways,” said Dr Chris Bishop, from Microsoft Research.

̽��ֱ��next steps for Aardvark include developing a new team within the Alan Turing Institute led by Turner, who will explore the potential to deploy Aardvark in the global south and integrate the technology into the Institute’s wider work to develop high-precision environmental forecasting for weather, oceans and sea ice.

Reference:
Anna Allen, Stratis Markou et al. ‘End-to-end data-driven weather prediction.’ Nature (2025). DOI: 10.1038/s41586-025-08897-0

Adapted from a media release by ̽��ֱ��Alan Turing Institute

A new AI weather prediction system, developed by researchers from the ̽��ֱ�� of Cambridge, can deliver accurate forecasts tens of times faster and using thousands of times less computing power than current AI and physics-based forecasting systems.

̽��ֱ��Alan Turing Institute

Professor Richard Turner using Aardvark Weather

̽��ֱ��text in this work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. Images, including our videos, are Copyright © ̽��ֱ�� of Cambridge and licensors/contributors as identified. All rights reserved. We make our image and video content available in a number of ways – on our main website under its Terms and conditions, and on a range of channels including social media that permit your use and sharing of our content under their respective Terms.

Yes

Under the bonnet at Dawn, the UK's fastest AI supercomputer

hcf38 — Mon, 17 Mar 2025 15:58:08 +0000

How AI supercomputer, Dawn, is being used to tackle some of the most pressing issues facing humanity.

Artificial intelligence outperforms clinical tests at predicting progress of Alzheimer’s disease

cjb250 — Fri, 12 Jul 2024 22:30:52 +0000

̽��ֱ��team say this new approach could reduce the need for invasive and costly diagnostic tests while improving treatment outcomes early when interventions such as lifestyle changes or new medicines may have a chance to work best.

Dementia poses a significant global healthcare challenge, affecting over 55 million people worldwide at an estimated annual cost of $820 billion. ̽��ֱ��number of cases is expected to almost treble over the next 50 years.

̽��ֱ��main cause of dementia is Alzheimer’s disease, which accounts for 60-80% of cases. Early detection is crucial as this is when treatments are likely to be most effective, yet early dementia diagnosis and prognosis may not be accurate without the use of invasive or expensive tests such as positron emission tomography (PET) scans or lumbar puncture, which are not available in all memory clinics. As a result, up to a third of patients may be misdiagnosed and others diagnosed too late for treatment to be effective.

A team led by scientists from the Department of Psychology at the ̽��ֱ�� of Cambridge has developed a machine learning model able to predict whether and how fast an individual with mild memory and thinking problems will progress to developing Alzheimer’s disease. In research published today in eClinical Medicine, they show that it is more accurate than current clinical diagnostic tools.

To build their model, the researchers used routinely-collected, non-invasive, and low-cost patient data – cognitive tests and structural MRI scans showing grey matter atrophy – from over 400 individuals who were part of a research cohort in the USA.

They then tested the model using real-world patient data from a further 600 participants from the US cohort and – importantly – longitudinal data from 900 people from memory clinics in the UK and Singapore.

̽��ֱ��algorithm was able to distinguish between people with stable mild cognitive impairment and those who progressed to Alzheimer’s disease within a three-year period. It was able to correctly identify individuals who went on to develop Alzheimer’s in 82% of cases and correctly identify those who didn’t in 81% of cases from cognitive tests and an MRI scan alone.

̽��ֱ��algorithm was around three times more accurate at predicting the progression to Alzheimer’s than the current standard of care; that is, standard clinical markers (such as grey matter atrophy or cognitive scores) or clinical diagnosis. This shows that the model could significantly reduce misdiagnosis.

̽��ֱ��model also allowed the researchers to stratify people with Alzheimer’s disease using data from each person’s first visit at the memory clinic into three groups: those whose symptoms would remain stable (around 50% of participants), those who would progress to Alzheimer’s slowly (around 35%) and those who would progress more rapidly (the remaining 15%). These predictions were validated when looking at follow-up data over 6 years. This is important as it could help identify those people at an early enough stage that they may benefit from new treatments, while also identifying those people who need close monitoring as their condition is likely to deteriorate rapidly.

Importantly, those 50% of people who have symptoms such as memory loss but remain stable, would be better directed to a different clinical pathway as their symptoms may be due to other causes rather than dementia, such as anxiety or depression.

Senior author Professor Zoe Kourtzi from the Department of Psychology at the ̽��ֱ�� of Cambridge said: “We’ve created a tool which, despite using only data from cognitive tests and MRI scans, is much more sensitive than current approaches at predicting whether someone will progress from mild symptoms to Alzheimer’s – and if so, whether this progress will be fast or slow.

“This has the potential to significantly improve patient wellbeing, showing us which people need closest care, while removing the anxiety for those patients we predict will remain stable. At a time of intense pressure on healthcare resources, this will also help remove the need for unnecessary invasive and costly diagnostic tests.”

While the researchers tested the algorithm on data from a research cohort, it was validated using independent data that included almost 900 individuals who attended memory clinics in the UK and Singapore. In the UK, patients were recruited through the Quantiative MRI in NHS Memory Clinics Study (QMIN-MC) led by study co-author Dr Timothy Rittman at Cambridge ̽��ֱ�� Hospitals NHS Trust and Cambridgeshire and Peterborough NHS Foundation Trusts (CPFT).

̽��ֱ��researchers say this shows it should be applicable in a real-world patient, clinical setting.

Dr Ben Underwood, Honorary Consultant Psychiatrist at CPFT and assistant professor at the Department of Psychiatry, ̽��ֱ�� of Cambridge, said: “Memory problems are common as we get older. In clinic I see how uncertainty about whether these might be the first signs of dementia can cause a lot of worry for people and their families, as well as being frustrating for doctors who would much prefer to give definitive answers. ̽��ֱ��fact that we might be able to reduce this uncertainty with information we already have is exciting and is likely to become even more important as new treatments emerge.”

Professor Kourtzi said: “AI models are only as good as the data they are trained on. To make sure ours has the potential to be adopted in a healthcare setting, we trained and tested it on routinely-collected data not just from research cohorts, but from patients in actual memory clinics. This shows it will be generalisable to a real-world setting.”

̽��ֱ��team now hope to extend their model to other forms of dementia, such as vascular dementia and frontotemporal dementia, and using different types of data, such as markers from blood tests.

Professor Kourtzi added: “If we’re going to tackle the growing health challenge presented by dementia, we will need better tools for identifying and intervening at the earliest possible stage. Our vision is to scale up our AI tool to help clinicians assign the right person at the right time to the right diagnostic and treatment pathway. Our tool can help match the right patients to clinical trials, accelerating new drug discovery for disease modifying treatments.”

This work was in collaboration with a cross-disciplinary team including Professor Peter Tino at the ̽��ֱ�� of Birmingham and Professor Christopher Chen at the National ̽��ֱ�� of Singapore. It was funded by Wellcome, the Royal Society, Alzheimer’s Research UK, the Alzheimer’s Drug Discovery Foundation Diagnostics Accelerator, the Alan Turing Institute, and the National Institute for Health and Care Research Cambridge Biomedical Research Centre.

Reference
Lee, LY & Vaghari, D et al. Robust and interpretable AI-guided marker for early dementia prediction in real-world clinical settings. eClinMed; 12 July 2024; DOI: 10.1016/j.eclinm.2024.102725

Cambridge scientists have developed an artificially-intelligent tool capable of predicting in four cases out of five whether people with early signs of dementia will remain stable or develop Alzheimer’s disease.

We’ve created a tool which is much more sensitive than current approaches at predicting whether someone will progress from mild symptoms to Alzheimer’s

Zoe Kourtzi

Yuichiro Chino (Getty Images)

Brain on molecular structure, circuitry, and programming code background

Yes

New open-source platform allows users to evaluate performance of AI-powered chatbots

sc604 — Tue, 04 Jun 2024 10:34:36 +0000

A team of computer scientists, engineers, mathematicians and cognitive scientists, led by the ̽��ֱ�� of Cambridge, developed an open-source evaluation platform called CheckMate, which allows human users to interact with and evaluate the performance of large language models (LLMs).

̽��ֱ��researchers tested CheckMate in an experiment where human participants used three LLMs – InstructGPT, ChatGPT and GPT-4 – as assistants for solving undergraduate-level mathematics problems.

̽��ֱ��team studied how well LLMs can assist participants in solving problems. Despite a generally positive correlation between a chatbot’s correctness and perceived helpfulness, the researchers also found instances where the LLMs were incorrect, but still useful for the participants. However, certain incorrect LLM outputs were thought to be correct by participants. This was most notable in LLMs optimised for chat.

̽��ֱ��researchers suggest models that communicate uncertainty, respond well to user corrections, and can provide a concise rationale for their recommendations, make better assistants. Human users of LLMs should verify their outputs carefully, given their current shortcomings.

̽��ֱ��results, reported in the Proceedings of the National Academy of Sciences (PNAS), could be useful in both informing AI literacy training, and help developers improve LLMs for a wider range of uses.

While LLMs are becoming increasingly powerful, they can also make mistakes and provide incorrect information, which could have negative consequences as these systems become more integrated into our everyday lives.

“LLMs have become wildly popular, and evaluating their performance in a quantitative way is important, but we also need to evaluate how well these systems work with and can support people,” said co-first author Albert Jiang, from Cambridge’s Department of Computer Science and Technology. “We don’t yet have comprehensive ways of evaluating an LLM’s performance when interacting with humans.”

̽��ֱ��standard way to evaluate LLMs relies on static pairs of inputs and outputs, which disregards the interactive nature of chatbots, and how that changes their usefulness in different scenarios. ̽��ֱ��researchers developed CheckMate to help answer these questions, designed for but not limited to applications in mathematics.

“When talking to mathematicians about LLMs, many of them fall into one of two main camps: either they think that LLMs can produce complex mathematical proofs on their own, or that LLMs are incapable of simple arithmetic,” said co-first author Katie Collins from the Department of Engineering. “Of course, the truth is probably somewhere in between, but we wanted to find a way of evaluating which tasks LLMs are suitable for and which they aren’t.”

̽��ֱ��researchers recruited 25 mathematicians, from undergraduate students to senior professors, to interact with three different LLMs (InstructGPT, ChatGPT, and GPT-4) and evaluate their performance using CheckMate. Participants worked through undergraduate-level mathematical theorems with the assistance of an LLM and were asked to rate each individual LLM response for correctness and helpfulness. Participants did not know which LLM they were interacting with.

̽��ֱ��researchers recorded the sorts of questions asked by participants, how participants reacted when they were presented with a fully or partially incorrect answer, whether and how they attempted to correct the LLM, or if they asked for clarification. Participants had varying levels of experience with writing effective prompts for LLMs, and this often affected the quality of responses that the LLMs provided.

An example of an effective prompt is “what is the definition of X” (X being a concept in the problem) as chatbots can be very good at retrieving concepts they know of and explaining it to the user.

“One of the things we found is the surprising fallibility of these models,” said Collins. “Sometimes, these LLMs will be really good at higher-level mathematics, and then they’ll fail at something far simpler. It shows that it’s vital to think carefully about how to use LLMs effectively and appropriately.”

However, like the LLMs, the human participants also made mistakes. ̽��ֱ��researchers asked participants to rate how confident they were in their own ability to solve the problem they were using the LLM for. In cases where the participant was less confident in their own abilities, they were more likely to rate incorrect generations by LLM as correct.

“This kind of gets to a big challenge of evaluating LLMs, because they’re getting so good at generating nice, seemingly correct natural language, that it’s easy to be fooled by their responses,” said Jiang. “It also shows that while human evaluation is useful and important, it’s nuanced, and sometimes it’s wrong. Anyone using an LLM, for any application, should always pay attention to the output and verify it themselves.”

Based on the results from CheckMate, the researchers say that newer generations of LLMs are increasingly able to collaborate helpfully and correctly with human users on undergraduate-level maths problems, as long as the user can assess the correctness of LLM-generated responses. Even if the answers may be memorised and can be found somewhere on the internet, LLMs have the advantage of being flexible in their inputs and outputs over traditional search engines (though should not replace search engines in their current form).

While CheckMate was tested on mathematical problems, the researchers say their platform could be adapted to a wide range of fields. In the future, this type of feedback could be incorporated into the LLMs themselves, although none of the CheckMate feedback from the current study has been fed back into the models.

“These kinds of tools can help the research community to have a better understanding of the strengths and weaknesses of these models,” said Collins. “We wouldn’t use them as tools to solve complex mathematical problems on their own, but they can be useful assistants if the users know how to take advantage of them.”

̽��ֱ��research was supported in part by the Marshall Commission, the Cambridge Trust, Peterhouse, Cambridge, ̽��ֱ��Alan Turing Institute, the European Research Council, and the Engineering and Physical Sciences Research Council (EPSRC), part of UK Research and Innovation (UKRI).

Reference:
Katherine M Collins, Albert Q Jiang, et al. ‘Evaluating Language Models for Mathematics through Interactions.’ Proceedings of the National Academy of Sciences (2024). DOI: 10.1073/pnas.2318124121

Researchers have developed a platform for the interactive evaluation of AI-powered chatbots such as ChatGPT.

Anyone using an LLM, for any application, should always pay attention to the output and verify it themselves

Albert Jiang

da-kuk via Getty Images

Chatbot

Yes

How sure is sure? Incorporating human error into machine learning

sc604 — Wed, 09 Aug 2023 23:40:58 +0000

Human error and uncertainty are concepts that many artificial intelligence systems fail to grasp, particularly in systems where a human provides feedback to a machine learning model. Many of these systems are programmed to assume that humans are always certain and correct, but real-world decision-making includes occasional mistakes and uncertainty.

Researchers from the ̽��ֱ�� of Cambridge, along with ̽��ֱ��Alan Turing Institute, Princeton, and Google DeepMind, have been attempting to bridge the gap between human behaviour and machine learning, so that uncertainty can be more fully accounted for in AI applications where humans and machines are working together. This could help reduce risk and improve trust and reliability of these applications, especially where safety is critical, such as medical diagnosis.

̽��ֱ��team adapted a well-known image classification dataset so that humans could provide feedback and indicate their level of uncertainty when labelling a particular image. ̽��ֱ��researchers found that training with uncertain labels can improve these systems’ performance in handling uncertain feedback, although humans also cause the overall performance of these hybrid systems to drop. Their results will be reported at the AAAI/ACM Conference on Artificial Intelligence, Ethics and Society (AIES 2023) in Montréal.

‘Human-in-the-loop’ machine learning systems – a type of AI system that enables human feedback – are often framed as a promising way to reduce risks in settings where automated models cannot be relied upon to make decisions alone. But what if the humans are unsure?

“Uncertainty is central in how humans reason about the world but many AI models fail to take this into account,” said first author Katherine Collins from Cambridge’s Department of Engineering. “A lot of developers are working to address model uncertainty, but less work has been done on addressing uncertainty from the person’s point of view.”

We are constantly making decisions based on the balance of probabilities, often without really thinking about it. Most of the time – for example, if we wave at someone who looks just like a friend but turns out to be a total stranger – there’s no harm if we get things wrong. However, in certain applications, uncertainty comes with real safety risks.

“Many human-AI systems assume that humans are always certain of their decisions, which isn’t how humans work – we all make mistakes,” said Collins. “We wanted to look at what happens when people express uncertainty, which is especially important in safety-critical settings, like a clinician working with a medical AI system.”

“We need better tools to recalibrate these models, so that the people working with them are empowered to say when they’re uncertain,” said co-author Matthew Barker, who recently completed his MEng degree at Gonville & Caius College, Cambridge. “Although machines can be trained with complete confidence, humans often can’t provide this, and machine learning models struggle with that uncertainty.”

For their study, the researchers used some of the benchmark machine learning datasets: one was for digit classification, another for classifying chest X-rays, and one for classifying images of birds. For the first two datasets, the researchers simulated uncertainty, but for the bird dataset, they had human participants indicate how certain they were of the images they were looking at: whether a bird was red or orange, for example. These annotated ‘soft labels’ provided by the human participants allowed the researchers to determine how the final output was changed. However, they found that performance degraded rapidly when machines were replaced with humans.

“We know from decades of behavioural research that humans are almost never 100% certain, but it’s a challenge to incorporate this into machine learning,” said Barker. “We’re trying to bridge the two fields so that machine learning can start to deal with human uncertainty where humans are part of the system.”

̽��ֱ��researchers say their results have identified several open challenges when incorporating humans into machine learning models. They are releasing their datasets so that further research can be carried out and uncertainty might be built into machine learning systems.

“As some of our colleagues so brilliantly put it, uncertainty is a form of transparency, and that’s hugely important,” said Collins. “We need to figure out when we can trust a model and when to trust a human and why. In certain applications, we’re looking at probability over possibilities. Especially with the rise of chatbots, for example, we need models that better incorporate the language of possibility, which may lead to a more natural, safe experience.”

“In some ways, this work raised more questions than it answered,” said Barker. “But even though humans may be miscalibrated in their uncertainty, we can improve the trustworthiness and reliability of these human-in-the-loop systems by accounting for human behaviour.”

̽��ֱ��research was supported in part by the Cambridge Trust, the Marshall Commission, the Leverhulme Trust, the Gates Cambridge Trust and the Engineering and Physical Sciences Research Council (EPSRC), part of UK Research and Innovation (UKRI).

Reference:
Katherine M Collins et al. ‘Human Uncertainty in Concept-Based AI Systems.’ Paper presented at the Sixth AAAI/ACM Conference on Artificial Intelligence, Ethics and Society (AIES 2023), August 8-10, 2023. Montréal, QC, Canada.

Researchers are developing a way to incorporate one of the most human of characteristics – uncertainty – into machine learning systems.

Uncertainty is central in how humans reason about the world but many AI models fail to take this into account

Katherine Collins

PeopleImages via Getty Images

Futuristic image of a doctor looking at brain scans

̽��ֱ��text in this work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. Images, including our videos, are Copyright © ̽��ֱ�� of Cambridge and licensors/contributors as identified. All rights reserved. We make our image and video content available in a number of ways – as here, on our main website under its Terms and conditions, and on a range of channels including social media that permit your use and sharing of our content under their respective Terms.

Yes

Machine learning algorithm predicts how to get the most out of electric vehicle batteries

sc604 — Tue, 23 Aug 2022 09:01:34 +0000

̽��ֱ��researchers, from the ̽��ֱ�� of Cambridge, say their algorithm could help drivers, manufacturers and businesses get the most out of the batteries that power electric vehicles by suggesting routes and driving patterns that minimise battery degradation and charging times.

̽��ֱ��team developed a non-invasive way to probe batteries and get a holistic view of battery health. These results were then fed into a machine learning algorithm that can predict how different driving patterns will affect the future health of the battery.

If developed commercially, the algorithm could be used to recommend routes that get drivers from point to point in the shortest time without degrading the battery, for example, or recommend the fastest way to charge the battery without causing it to degrade. ̽��ֱ��results are reported in the journal Nature Communications.

̽��ֱ��health of a battery, whether it’s in a smartphone or a car, is far more complex than a single number on a screen. “Battery health, like human health, is a multi-dimensional thing, and it can degrade in lots of different ways,” said first author Penelope Jones, from Cambridge’s Cavendish Laboratory. “Most methods of monitoring battery health assume that a battery is always used in the same way. But that’s not how we use batteries in real life. If I’m streaming a TV show on my phone, it’s going to run down the battery a whole lot faster than if I’m using it for messaging. It’s the same with electric cars – how you drive will affect how the battery degrades.”

“Most of us will replace our phones well before the battery degrades to the point that it’s unusable, but for cars, the batteries need to last for five, ten years or more,” said Dr Alpha Lee, who led the research. “Battery capacity can change drastically over that time, so we wanted to come up with a better way of checking battery health.”

̽��ֱ��researchers developed a non-invasive probe that sends high-dimensional electrical pulses into a battery and measures the response, providing a series of ‘biomarkers’ of battery health. This method is gentle on the battery and doesn’t cause it to degrade any further.

̽��ֱ��electrical signals from the battery were converted into a description of the battery’s state, which was fed into a machine learning algorithm. ̽��ֱ��algorithm was able to predict how the battery would respond in the next charge-discharge cycle, depending on how quickly the battery was charged and how fast the car would be going the next time it was on the road. Tests with 88 commercial batteries showed that the algorithm did not require any information about previous usage of the battery to make an accurate prediction.

̽��ֱ��experiment focused on lithium cobalt oxide (LCO) cells, which are widely used in rechargeable batteries, but the method is generalisable across the different types of battery chemistries used in electric vehicles today.

“This method could unlock value in so many parts of the supply chain, whether you’re a manufacturer, an end user, or a recycler, because it allows us to capture the health of the battery beyond a single number, and because it’s predictive,” said Lee. “It could reduce the time it takes to develop new types of batteries, because we’ll be able to predict how they will degrade under different operating conditions.”

̽��ֱ��researchers say that in addition to manufacturers and drivers, their method could be useful for businesses that operate large fleets of electric vehicles, such as logistics companies. “ ̽��ֱ��framework we’ve developed could help companies optimise how they use their vehicles to improve the overall battery life of the fleet,” said Lee. “There’s so much potential with a framework like this.”

“It’s been such an exciting framework to build because it could solve so many of the challenges in the battery field today,” said Jones. “It’s a great time to be involved in the field of battery research, which is so important in helping address climate change by transitioning away from fossil fuels.”

̽��ֱ��researchers are now working with battery manufacturers to accelerate the development of safer, longer-lasting next-generation batteries. They are also exploring how their framework could be used to develop optimal fast charging protocols to reduce electric vehicle charging times without causing degradation.

̽��ֱ��research was supported by the Winton Programme for the Physics of Sustainability, the Ernest Oppenheimer Fund, ̽��ֱ��Alan Turing Institute and the Royal Society.

Reference:
Penelope K Jones, Ulrich Stimming & Alpha A Lee. ‘Impedance-based forecasting of lithium-ion battery performance amid uneven usage.’ Nature Communications (2022). DOI: 10.1038/s41467-022-32422-w

For more information on energy-related research in Cambridge, please visit Energy IRC, which brings together Cambridge’s research knowledge and expertise, in collaboration with global partners, to create solutions for a sustainable and resilient energy landscape for generations to come.

Researchers have developed a machine learning algorithm that could help reduce charging times and prolong battery life in electric vehicles by predicting how different driving patterns affect battery performance, improving safety and reliability.

This method could unlock value in so many parts of the supply chain, whether you’re a manufacturer, an end user, or a recycler, because it allows us to capture the health of the battery beyond a single number

Alpha Lee

Monty Rakusen via Getty Images

People charging their electric cars at charging station in York

̽��ֱ��text in this work is licensed under a Creative Commons Attribution 4.0 International License. Images, including our videos, are Copyright © ̽��ֱ�� of Cambridge and licensors/contributors as identified. All rights reserved. We make our image and video content available in a number of ways – as here, on our main website under its Terms and conditions, and on a range of channels including social media that permit your use and sharing of our content under their respective Terms.

Yes

‘Transformational’ approach to machine learning could accelerate search for new disease treatments

sc604 — Mon, 29 Nov 2021 20:00:00 +0000

̽��ֱ��method, called transformational machine learning (TML), was developed by a team from the UK, Sweden, India and Netherlands. It learns from multiple problems and improves performance while it learns.

TML could accelerate the identification and production of new drugs by improving the machine learning systems which are used to identify them. ̽��ֱ��results are reported in the Proceedings of the National Academy of Sciences.

Most types of machine learning (ML) use labelled examples, and these examples are almost always represented in the computer using intrinsic features, such as the colour or shape of an object. ̽��ֱ��computer then forms general rules that relate the features to the labels.

“It’s sort of like teaching a child to identify different animals: this is a rabbit, this is a donkey and so on,” said Professor Ross King from Cambridge’s Department of Chemical Engineering and Biotechnology, who led the research. “If you teach a machine learning algorithm what a rabbit looks like, it will be able to tell whether an animal is or isn’t a rabbit. This is the way that most machine learning works – it deals with problems one at a time.”

However, this is not the way that human learning works: instead of dealing with a single issue at a time, we get better at learning because we have learned things in the past.

“To develop TML, we applied this approach to machine learning, and developed a system that learns information from previous problems it has encountered in order to better learn new problems,” said King, who is also a Fellow at ̽��ֱ��Alan Turing Institute. “Where a typical ML system has to start from scratch when learning to identify a new type of animal - say a kitten - TML can use the similarity to existing animals: kittens are cute like rabbits, but don’t have long ears like rabbits and donkeys. This makes TML a much more powerful approach to machine learning.”

̽��ֱ��researchers demonstrated the effectiveness of their idea on thousands of problems from across science and engineering. They say it shows particular promise in the area of drug discovery, where this approach speeds up the process by checking what other ML models say about a particular molecule. A typical ML approach will search for drug molecules of a particular shape, for example. TML instead uses the connection of the drugs to other drug discovery problems.

“I was surprised how well it works – better than anything else we know for drug design,” said King. “It’s better at choosing drugs than humans are – and without the best science, we won’t get the best results.”

Reference:
Ivan Olier et al. ‘Transformational Machine Learning: Learning How to Learn from Many Related Scientific Problems.’ Proceedings of the National Academy of Sciences (2021). DOI: 10.1073/pnas.2108013118

Researchers have developed a new approach to machine learning that ‘learns how to learn’ and out-performs current machine learning methods for drug design, which in turn could accelerate the search for new disease treatments.

I was surprised how well it works – better than anything else we know for drug design

Ross King

mahdis mousavi via Unsplash

Woman in grey shirt

Yes

AI could detect dementia years before symptoms appear

cjb250 — Thu, 12 Aug 2021 08:44:50 +0000

Artificial intelligence could spot the early signs of dementia from a simple brain scan long before major symptoms appear – and in some cases before any symptoms appear – say Cambridge researchers.