Meta tags:
description=Successfully controlling the nuclear fusion plasma in a tokamak with deep reinforcement learning;

Headings (most frequently used words):

ai, our, to, and, the, of, is, fusion, explore, next, generation, systems, latest, breakthroughs, updates, from, lab, unlocking, new, era, discovery, with, mission, build, responsibly, benefit, humanity, accelerating, science, through, learned, plasma, control, learning, when, data, hard, acquire, success, by, prioritising, simplicity, flexibility, future, beyond,

Text of the page (most frequently used words):
the (98), and (68), our (32), #plasma (32), for (27), gemini (25), with (20), google (16), control (15), #science (14), tokamak (14), explore (14), deepmind (13), that (13), research (12), fusion (12), models (11), build (10), learning (10), learn (9), more (9), breakthroughs (9), audio (9), new (9), this (9), controllers (8), from (8), tcv (8), generate (8), about (7), through (7), use (7), are (7), tools (7), updates (6), torax (6), reinforcement (6), these (6), real (6), applications (6), energy (6), successfully (6), example (6), system (6), can (6), nuclear (6), create (6), responsibility (5), safety (5), robotics (5), genie (5), latest (5), deep (5), scientific (5), design (5), could (5), shapes (5), epfl (5), many (5), vessel (5), next (5), controller (5), coils (5), high (5), frontier (4), humanity (4), conditions (4), used (4), fast (4), magnetic (4), plasmas (4), discover (4), world (4), predict (4), then (4), shape (4), different (4), experiments (4), single (4), first (4), time (4), have (4), how (4), using (4), perceive (4), reason (4), interact (4), interactive (4), worlds (4), anything (4), products (3), institute (3), podcast (3), accelerator (3), programs (3), national (3), partnerships (3), careers (3), news (3), alphaevolve (3), alphaearth (3), weathernext (3), alphagenome (3), alphafold (3), publications (3), evals (3), veo (3), lyria (3), gemma (3), nano (3), banana (3), omni (3), responsibly (3), benefit (3), paper (3), simulator (3), train (3), advanced (3), computing (3), accuracy (3), open (3), accelerate (3), autonomously (3), tokamaks (3), work (3), points (3), complex (3), machines (3), human (3), hard (3), properties (3), under (3), help (3), walls (3), generation (3), was (3), even (3), there (3), two (3), all (3), video (3), experiment (3), you (3), systems (3), algorithms (3), which (3), researchers (3), advance (3), powerful (3), into (3), one (3), discovery (3), weather (3), forecasting (3), solve (3), mission (3), try (3), terms (2), privacy (2), antigravity (2), studio (2), app (2), sign (2), will (2), read (2), jax (2), may (2), 2024 (2), core (2), current (2), ability (2), exciting (2), capabilities (2), making (2), available (2), access (2), development (2), source (2), progress (2), when (2), applying (2), demonstration (2), shows (2), while (2), simultaneously (2), their (2), also (2), future (2), fields (2), where (2), problems (2), years (2), come (2), produced (2), range (2), physicists (2), credit (2), spc (2), made (2), legs (2), demonstrated (2), iter (2), had (2), never (2), before (2), inside (2), able (2), requested (2), see (2), top (2), would (2), multiple (2), requiring (2), each (2), uses (2), adjust (2), voltage (2), architecture (2), configuration (2), simulators (2), simulation (2), sculpt (2), way (2), second (2), day (2), develop (2), both (2), simulated (2), only (2), needs (2), share (2), chemistry (2), mathematics (2), ensure (2), has (2), avenues (2), scientists (2), published (2), hydrogen (2), process (2), contain (2), 2022 (2), accelerating (2), learned (2), looking (2), people (2), who (2), want (2), make (2), positive (2), impact (2), projects (2), ensuring (2), proactive (2), security (2), against (2), evolving (2), threats (2), education (2), skills (2), experimental (2), math (2), map (2), planet (2), unprecedented (2), detail (2), accurate (2), understand (2), genome (2), protein (2), structures (2), unlocking (2), era (2), mastering (2), game (2), alphago (2), agent (2), plays (2), reasons (2), learns (2), sima (2), lab (2), responsible (2), scale (2), fidelity (2), music (2), cinematic (2), specialized (2), talk (2), edit (2), detailed (2), images (2), intelligent (2), agents (2), cookies, management, controls, accept, acknowledge, information, accordance, policy, innovations, follow, differentiable, transport, notes, released, interior, predicts, changes, temperature, density, electric, expands, written, python, framework, originally, developed, offers, scalable, computation, increased, prediction, sensitivity, analysis, community, hope, enables, workflows, general, purpose, optimization, similar, seen, other, domains, successful, power, assist, expect, increasing, sophistication, going, forward, capability, creating, kinds, designing, bright, especially, consider, augment, expertise, serving, tool, creative, approaches, transformative, technology, industrial, ranging, efficiency, personalised, medicine, beyond, whose, study, created, being, studied, usefulness, generating, snowflake, reduce, cost, cooling, spreading, exhaust, contact, close, proposal, construction, conducting, behaviour, did, something, been, done, stabilising, droplet, find, simply, changed, goal, algorithm, found, appropriate, above, instant, takes, according, shifts, downward, detaches, suspending, middle, held, stationary, needed, measure, finally, steered, back, safely, destroyed, trained, steers, phases, left, view, during, right, reconstructed, target, wanted, hit, showed, manipulate, aspects, existing, separate, estimate, magnets, accordingly, contrast, neural, network, once, automatically, voltages, best, achieve, directly, sensors, success, prioritising, simplicity, flexibility, given, obstacles, turned, partners, built, set, model, dynamics, were, allow, validate, results, showing, desired, whilst, cheaper, convenient, still, overcome, barriers, slow, require, hours, computer, simulate, addition, condition, change, algorithmic, improvements, physical, adapt, realities, hardware, currently, limited, run, dozens, active, around, they, expensive, demand, sustain, three, seconds, after, minutes, cool, down, reset, attempt, not, groups, often, further, limiting, data, acquire, another, machine, expert, communities, together, tackle, grand, challenges, team, approach, diverse, quantum, pure, material, fundamental, benefits, founder, ceo, demis, hassabis, last, potential, unlock, entirely, across, biology, now, physics, describe, building, running, variable, lausanne, switzerland, combines, environment, keep, steady, accurately, sculpting, controlled, superheated, matter, importantly, allows, investigate, reacts, improving, understanding, reactors, today, nature, global, crisis, long, sought, clean, limitless, reaction, powers, stars, universe, contender, smashing, fusing, common, element, seawater, releases, huge, amounts, here, earth, recreated, extreme, doughnut, shaped, vacuum, surrounded, hotter, than, sun, however, inherently, unstable, sustaining, required, challenge, coordinate, them, thousands, times, per, touches, result, heat, loss, possibly, damage, problem, part, collaborated, opening, swiss, center, note, blog, feb, following, release, minor, text, reflect, code, controlling, copied, february, agentic, platform, gen, chat, apps, labs, skip, main, content,

Text of the page (random words):
accelerating fusion science through learned plasma control google deepmind skip to main content explore our next generation ai systems explore models gemini gemini build intelligent agents gemini omni create anything from anything nano banana create and edit detailed images gemini audio talk create and control audio specialized models veo generate cinematic video with audio lyria generate high fidelity music and audio genie 3 generate and explore interactive worlds gemini robotics perceive reason use tools and interact open models gemma build responsible ai applications at scale our latest ai breakthroughs and updates from the lab explore research breakthroughs sima 2 an agent that plays reasons and learns with you genie 3 generate and explore interactive worlds alphago mastering the game of go gemini robotics perceive reason use tools and interact learn more evals publications responsibility frontier safety unlocking a new era of discovery with ai explore science breakthroughs alphafold predict protein structures with high accuracy alphagenome using ai to understand the human genome weathernext fast and accurate ai weather forecasting alphaearth map our planet in unprecedented detail alphaevolve design advanced algorithms for math and applications in computing learn more gemini for science experimental tools science skills our mission is to build ai responsibly to benefit humanity about google deepmind responsibility ensuring ai safety through proactive security even against evolving threats news discover our latest ai breakthroughs projects and updates careers we re looking for people who want to make a real positive impact on the world learn more education our national partnerships for ai accelerator programs the podcast deepmind institute models explore our next generation ai systems explore models gemini gemini build intelligent agents gemini omni create anything from anything nano banana create and edit detailed images gemini audio talk create and control audio specialized models veo generate cinematic video with audio lyria generate high fidelity music and audio genie 3 generate and explore interactive worlds gemini robotics perceive reason use tools and interact open models gemma build responsible ai applications at scale research our latest ai breakthroughs and updates from the lab explore research breakthroughs sima 2 an agent that plays reasons and learns with you genie 3 generate and explore interactive worlds alphago mastering the game of go gemini robotics perceive reason use tools and interact learn more evals publications responsibility frontier safety science unlocking a new era of discovery with ai explore science breakthroughs alphafold predict protein structures with high accuracy alphagenome using ai to understand the human genome weathernext fast and accurate ai weather forecasting alphaearth map our planet in unprecedented detail alphaevolve design advanced algorithms for math and applications in computing learn more gemini for science experimental tools science skills about our mission is to build ai responsibly to benefit humanity about google deepmind learn more education our national partnerships for ai accelerator programs the podcast deepmind institute responsibility ensuring ai safety through proactive security even against evolving threats news discover our latest ai breakthroughs projects and updates careers we re looking for people who want to make a real positive impact on the world build with gemini try gemini google deepmind google ai learn about all our ai google deepmind explore the frontier of ai google labs try our ai experiments google research explore our research products and apps gemini app chat with gemini google ai studio build with our next gen ai models google antigravity our agentic development platform models research science about build with gemini try gemini february 16 2022 science accelerating fusion science through learned plasma control share copied successfully controlling the nuclear fusion plasma in a tokamak with deep reinforcement learning note this blog was first published on 16 feb 2022 following the release of torax plasma simulator code in may 2024 we ve made minor updates to the text to reflect this to solve the global energy crisis researchers have long sought a source of clean limitless energy nuclear fusion the reaction that powers the stars of the universe is one contender by smashing and fusing hydrogen a common element of seawater the powerful process releases huge amounts of energy here on earth one way scientists have recreated these extreme conditions is by using a tokamak a doughnut shaped vacuum surrounded by magnetic coils that is used to contain a plasma of hydrogen that is hotter than the core of the sun however the plasmas in these machines are inherently unstable making sustaining the process required for nuclear fusion a complex challenge for example a control system needs to coordinate the tokamak s many magnetic coils and adjust the voltage on them thousands of times per second to ensure the plasma never touches the walls of the vessel which would result in heat loss and possibly damage to help solve this problem and as part of deepmind s mission to advance science we collaborated with the swiss plasma center at epfl to develop the first deep reinforcement learning rl system to autonomously discover how to control these coils and successfully contain the plasma in a tokamak opening new avenues to advance nuclear fusion research in a paper published today in nature we describe how we can successfully control nuclear fusion plasma by building and running controllers on the variable configuration tokamak tcv in lausanne switzerland using a learning architecture that combines deep rl and a simulated environment we produced controllers that can both keep the plasma steady and be used to accurately sculpt it into different shapes this plasma sculpting shows the rl system has successfully controlled the superheated matter and importantly allows scientists to investigate how the plasma reacts under different conditions improving our understanding of fusion reactors in the last two years deepmind has demonstrated ai s potential to accelerate scientific progress and unlock entirely new avenues of research across biology chemistry mathematics and now physics demis hassabis co founder and ceo deepmind this work is another powerful example of how machine learning and expert communities can come together to tackle grand challenges and accelerate scientific discovery our team is hard at work applying this approach to fields as diverse as quantum chemistry pure mathematics material design weather forecasting and more to solve fundamental problems and ensure ai benefits humanity learning when data is hard to acquire research into nuclear fusion is currently limited by researchers ability to run experiments while there are dozens of active tokamaks around the world they re expensive machines and in high demand for example tcv can only sustain the plasma in a single experiment for up to three seconds after which it needs 15 minutes to cool down and reset before the next attempt not only that multiple research groups often share use of the tokamak further limiting the time available for experiments given the current obstacles to access a tokamak researchers have turned to simulators to help advance research for example our partners at epfl have built a powerful set of simulation tools that model the dynamics of tokamaks we were able to use these to allow our rl system to learn to control tcv in simulation and then validate our results on the real tcv showing we could successfully sculpt the plasma into the desired shapes whilst this is a cheaper and more convenient way to train our controllers we still had to overcome many barriers for example plasma simulators are slow and require many hours of computer time to simulate one second of real time in addition the condition of tcv can change from day to day requiring us to develop algorithmic improvements both physical and simulated and to adapt to the realities of the hardware success by prioritising simplicity and flexibility existing plasma control systems are complex requiring separate controllers for each of tcv s 19 magnetic coils each controller uses algorithms to estimate the properties of the plasma in real time and adjust the voltage of the magnets accordingly in contrast our architecture uses a single neural network to control all of the coils at once automatically learning which voltages are the best to achieve a plasma configuration directly from sensors as a demonstration we first showed that we could manipulate many aspects of the plasma with a single controller the controller trained with deep reinforcement learning steers the plasma through multiple phases of an experiment on the left there is an inside view in the tokamak during the experiment on the right you can see the reconstructed plasma shape and the target points we wanted to hit credit deepmind spc epfl in the video above we see the plasma at the top of tcv at the instant our system takes control our controller first shapes the plasma according to the requested shape then shifts the plasma downward and detaches it from the walls suspending it in the middle of the vessel on two legs the plasma is held stationary as would be needed to measure plasma properties then finally the plasma is steered back to the top of the vessel and safely destroyed we then created a range of plasma shapes being studied by plasma physicists for their usefulness in generating energy for example we made a snowflake shape with many legs that could help reduce the cost of cooling by spreading the exhaust energy to different contact points on the vessel walls we also demonstrated a shape close to the proposal for iter the next generation tokamak under construction as epfl was conducting experiments to predict the behaviour of plasmas in iter we even did something that had never been done in tcv before by stabilising a droplet where there are two plasmas inside the vessel simultaneously our single system was able to find controllers for all of these different conditions we simply changed the goal we requested and our algorithm autonomously found an appropriate controller we successfully produced a range of shapes whose properties are under study by plasma physicists credit deepmind spc epfl the future of fusion and beyond similar to progress we ve seen when applying ai to other scientific domains our successful demonstration of tokamak control shows the power of ai to accelerate and assist fusion science and we expect increasing sophistication in the use of ai going forward this capability of autonomously creating controllers could be used to design new kinds of tokamaks while simultaneously designing their controllers our work also points to a bright future for reinforcement learning in the control of complex machines it s especially exciting to consider fields where ai could augment human expertise serving as a tool to discover new and creative approaches for hard real world problems we predict reinforcement learning will be a transformative technology for industrial and scientific control applications in the years to come with applications ranging from energy efficiency to personalised medicine in may 2024 we released torax a new open source plasma simulator torax models the core interior of the plasma and predicts changes in temperature density and electric current this expands our ability to train advanced tokamak ai controllers torax is written in jax a python framework originally developed to train ai and offers exciting capabilities for scientific computing through fast and scalable computation increased prediction accuracy and sensitivity analysis by making torax available to the fusion community with access to these new capabilities we hope it enables development of new workflows for general purpose tokamak design and optimization notes read the paper magnetic control of tokamak plasmas through deep reinforcement learning read the paper torax a fast and differentiable tokamak transport simulator in jax follow us sign up for updates on our latest innovations i accept google s terms and conditions and acknowledge that my information will be used in accordance with google s privacy policy sign up build ai responsibly to benefit humanity models gemini gemini omni nano banana gemini audio gemma genie lyria veo research gemini robotics breakthroughs evals publications frontier safety responsibility science alphafold alphagenome weathernext alphaearth alphaevolve products gemini app google ai studio google antigravity learn more about news careers national partnerships for ai accelerator programs the podcast deepmind institute about google google products privacy terms cookies management controls