Computational model discovery with #ReinforcementLearning
The motivation of this study is to leverage recent breakthroughs in artificial intelligence research to unlock novel solutions to important scientific problems encountered in computational science. To address the human intelligence limitations in discovering reduced-order models, we propose to supplement human thinking with artificial intelligence. Our three-pronged strategy consists of learning (i) models expressed in analytical form, (ii) which are evaluated a posteriori, and iii) using exclusively integral quantities from the reference solution as prior knowledge. In point (i), we pursue interpretable models expressed symbolically as opposed to black-box neural networks, the latter only being used during learning to efficiently parameterize the large search space of possible models. In point (ii), learned models are dynamically evaluated a posteriori in the computational solver instead of based on a priori information from preprocessed high-fidelity data, thereby accounting for the specificity of the solver at hand such as its numerics. Finally in point (iii), the exploration of new models is solely guided by predefined integral quantities, e.g., averaged quantities of engineering interest in Reynolds-averaged or large-eddy simulations (LES). We use a coupled deep reinforcement learning framework and computational solver to concurrently achieve these objectives. The combination of reinforcement learning with objectives (i), (ii) and (iii) differentiate our work from previous modeling attempts based on machine learning. In this report, we provide a high-level denoscription of the model discovery framework with reinforcement learning. The method is detailed for the application of discovering missing terms in differential equations. An elementary instantiation of the method is described that discovers missing terms in the Burgers' equation.
Paper
🔭 @DeepGravity
The motivation of this study is to leverage recent breakthroughs in artificial intelligence research to unlock novel solutions to important scientific problems encountered in computational science. To address the human intelligence limitations in discovering reduced-order models, we propose to supplement human thinking with artificial intelligence. Our three-pronged strategy consists of learning (i) models expressed in analytical form, (ii) which are evaluated a posteriori, and iii) using exclusively integral quantities from the reference solution as prior knowledge. In point (i), we pursue interpretable models expressed symbolically as opposed to black-box neural networks, the latter only being used during learning to efficiently parameterize the large search space of possible models. In point (ii), learned models are dynamically evaluated a posteriori in the computational solver instead of based on a priori information from preprocessed high-fidelity data, thereby accounting for the specificity of the solver at hand such as its numerics. Finally in point (iii), the exploration of new models is solely guided by predefined integral quantities, e.g., averaged quantities of engineering interest in Reynolds-averaged or large-eddy simulations (LES). We use a coupled deep reinforcement learning framework and computational solver to concurrently achieve these objectives. The combination of reinforcement learning with objectives (i), (ii) and (iii) differentiate our work from previous modeling attempts based on machine learning. In this report, we provide a high-level denoscription of the model discovery framework with reinforcement learning. The method is detailed for the application of discovering missing terms in differential equations. An elementary instantiation of the method is described that discovers missing terms in the Burgers' equation.
Paper
🔭 @DeepGravity
A deep-learning technique for phase identification in multiphase inorganic compounds using synthetic XRD powder patterns
Abstract
Here we report a facile, prompt protocol based on deep-learning techniques to sort out intricate phase identification and quantification problems in complex multiphase inorganic compounds. We simulate plausible powder X-ray powder diffraction (XRD) patterns for 170 inorganic compounds in the Sr-Li-Al-O quaternary compositional pool, wherein promising LED phosphors have been recently discovered. Finally, 1,785,405 synthetic XRD patterns are prepared by combinatorically mixing the simulated powder XRD patterns of 170 inorganic compounds. Convolutional neural network (CNN) models are built and eventually trained using this large prepared dataset. The fully trained CNN model promptly and accurately identifies the constituent phases in complex multiphase inorganic compounds. Although the CNN is trained using the simulated XRD data, a test with real experimental XRD data returns an accuracy of nearly 100% for phase identification and 86% for three-step-phase-fraction quantification.
Paper
🔭 @DeepGravity
Abstract
Here we report a facile, prompt protocol based on deep-learning techniques to sort out intricate phase identification and quantification problems in complex multiphase inorganic compounds. We simulate plausible powder X-ray powder diffraction (XRD) patterns for 170 inorganic compounds in the Sr-Li-Al-O quaternary compositional pool, wherein promising LED phosphors have been recently discovered. Finally, 1,785,405 synthetic XRD patterns are prepared by combinatorically mixing the simulated powder XRD patterns of 170 inorganic compounds. Convolutional neural network (CNN) models are built and eventually trained using this large prepared dataset. The fully trained CNN model promptly and accurately identifies the constituent phases in complex multiphase inorganic compounds. Although the CNN is trained using the simulated XRD data, a test with real experimental XRD data returns an accuracy of nearly 100% for phase identification and 86% for three-step-phase-fraction quantification.
Paper
🔭 @DeepGravity
Nature
A deep-learning technique for phase identification in multiphase inorganic compounds using synthetic XRD powder patterns
Nature Communications - Identifying the composition of multiphase inorganic compounds from XRD patterns is challenging. Here the authors use a convolutional neural network to identify phases in...
تقریبا در همه جا عکسها و اسمهایی منتشر شده است، اما هنوز شهامت آن را پیدا نکردهام که لیست کسانی که پرگشودند را بخوانم.
گویی این دومینوی درد و مرگ دیگر گوشش بدهکار تسلیت و نیایشمان نیست. بیدرنگ و بیتردید غرق خویشتن خویش است. آنچه که برای اوست شاید انجام بدون چون چرای تکلیفش باشد، لیک برای ما جانهای عزیزی است که ستانده میشود هر روز، و چکههای اشک بی وقفهی ماست که نهر میشود هر روز. برای او شاید اقتضای طبیعتش باشد، اما برای ما دردها و اندوههایی است که تا ژرفای دلهای سردرگممان رخنه میکند، شگفتزدگی و آلام بیانتهایی است که یارای شکیباییمان را میرباید.
اگرچه ابراز همدردی هیچ سفرکردهای را نیروی بازگشت نشده است، اما شاید ذرهای دلگرمی برای بازماندگان باشد. پس به رسم ادب، همدردی - این کوچکترین نماد احساس اندوه عمیق - به پیشگاه شما ابراز میشود.
به امید فردایی بهتر، شادتر و آزادتر برای ایران و ایرانی ...
@Reza
🔭 @DeepGravity
گویی این دومینوی درد و مرگ دیگر گوشش بدهکار تسلیت و نیایشمان نیست. بیدرنگ و بیتردید غرق خویشتن خویش است. آنچه که برای اوست شاید انجام بدون چون چرای تکلیفش باشد، لیک برای ما جانهای عزیزی است که ستانده میشود هر روز، و چکههای اشک بی وقفهی ماست که نهر میشود هر روز. برای او شاید اقتضای طبیعتش باشد، اما برای ما دردها و اندوههایی است که تا ژرفای دلهای سردرگممان رخنه میکند، شگفتزدگی و آلام بیانتهایی است که یارای شکیباییمان را میرباید.
اگرچه ابراز همدردی هیچ سفرکردهای را نیروی بازگشت نشده است، اما شاید ذرهای دلگرمی برای بازماندگان باشد. پس به رسم ادب، همدردی - این کوچکترین نماد احساس اندوه عمیق - به پیشگاه شما ابراز میشود.
به امید فردایی بهتر، شادتر و آزادتر برای ایران و ایرانی ...
@Reza
🔭 @DeepGravity
Computational Engineering Division Student Intern
Multiple Lecturer/Senior Lecturer (equivalent to U.S. Assistant Professor tenure track) positions in machine learning and computer vision at the University of Melbourne’s School of Computing and Information Systems.
MS/PhD/Visiting scholar positions for Deep RL Ford-WVU
PhD Positions for Robot Learning Uni Freiburg
2 PhD positions available – University College Cork (UCC)
Machine Learning Researcher, UK
New Post-doc Opening at U. of Toronto on Deep Learning / RL for Traffic Prediction and Control
RL/LfD research positions (including interns) at Bosch / UT Austin, focusing on autonomous vehicles
#Job
🔭 @DeepGravity
Multiple Lecturer/Senior Lecturer (equivalent to U.S. Assistant Professor tenure track) positions in machine learning and computer vision at the University of Melbourne’s School of Computing and Information Systems.
MS/PhD/Visiting scholar positions for Deep RL Ford-WVU
PhD Positions for Robot Learning Uni Freiburg
2 PhD positions available – University College Cork (UCC)
Machine Learning Researcher, UK
New Post-doc Opening at U. of Toronto on Deep Learning / RL for Traffic Prediction and Control
RL/LfD research positions (including interns) at Bosch / UT Austin, focusing on autonomous vehicles
#Job
🔭 @DeepGravity
Introducing neuromodulation in deep neural networks to learn adaptive behaviours
Abstract
Animals excel at adapting their intentions, attention, and actions to the environment, making them remarkably efficient at interacting with a rich, unpredictable and ever-changing external world, a property that intelligent machines currently lack. Such an adaptation property relies heavily on cellular neuromodulation, the biological mechanism that dynamically controls intrinsic properties of neurons and their response to external stimuli in a context-dependent manner. In this paper, we take inspiration from cellular neuromodulation to construct a new deep neural network architecture that is specifically designed to learn adaptive behaviours. The network adaptation capabilities are tested on navigation benchmarks in a meta-reinforcement learning context and compared with state-of-the-art approaches. Results show that neuromodulation is capable of adapting an agent to different tasks and that neuromodulation-based approaches provide a promising way of improving adaptation of artificial systems.
Paper
🔭 @DeepGravity
Abstract
Animals excel at adapting their intentions, attention, and actions to the environment, making them remarkably efficient at interacting with a rich, unpredictable and ever-changing external world, a property that intelligent machines currently lack. Such an adaptation property relies heavily on cellular neuromodulation, the biological mechanism that dynamically controls intrinsic properties of neurons and their response to external stimuli in a context-dependent manner. In this paper, we take inspiration from cellular neuromodulation to construct a new deep neural network architecture that is specifically designed to learn adaptive behaviours. The network adaptation capabilities are tested on navigation benchmarks in a meta-reinforcement learning context and compared with state-of-the-art approaches. Results show that neuromodulation is capable of adapting an agent to different tasks and that neuromodulation-based approaches provide a promising way of improving adaptation of artificial systems.
Paper
🔭 @DeepGravity
journals.plos.org
Introducing neuromodulation in deep neural networks to learn adaptive behaviours
Animals excel at adapting their intentions, attention, and actions to the environment, making them remarkably efficient at interacting with a rich, unpredictable and ever-changing external world, a property that intelligent machines currently lack. Such an…