The last decade has shown a tremendous success in solving various computer vision problems with the help of deep learning techniques. Lately, many works have demonstrated that learning-based approaches with suitable network architectures even exhibit superior performance for the solution of (ill-posed) image reconstruction problems such as deblurring, super-resolution, or medical image reconstruction. The drawback of purely learning-based methods, however, is that they cannot provide provable guarantees for the trained network to follow a given data formation process during inference. In this work we propose energy dissipating networks that iteratively compute a descent direction with respect to a given cost function or energy at the currently estimated reconstruction. Therefore, an adaptive step size rule such as a line-search, along with a suitable number of iterations can guarantee the reconstruction to follow a given data formation model encoded in the energy to arbitrary precision, and hence control the model's behavior even during test time. We prove that under standard assumptions, descent using the direction predicted by the network converges (linearly) to the global minimum of the energy. We illustrate the effectiveness of the proposed approach in experiments on single image super resolution and computed tomography (CT) reconstruction, and further illustrate extensions to convex feasibility problems.
Dynamic Modeling and Adaptive Controlling in GPS-Intelligent Buoy (GIB) Systems Based on Neural-Fuzzy Networks
School of Forestry, Henan Agricultural University(河南农业大学林学院);University of Malaya(马来亚大学);School of Management, Henan University of Technology(河南工业大学管理学院);Islamic Azad University(伊斯兰阿扎德大学);Obuda University(欧布达大学);Oxford Brookes University(牛津布鲁克斯大学)
Mathematical Institute, Faculty of Mathematics and Physics, Charles University(查理大学数学学院);Department of Mechanical Engineering and Applied Mechanics, University of Pennsylvania(宾夕法尼亚大学机械工程与应用力学系)
Graph Neural Network-Based Reinforcement Learning for Controlling Biological Networks - the GATTACA Framework
Faculty of Mathematics, Informatics and Mechanics, University of Warsaw(数学、信息学与力学学院,华沙大学);IDEAS Research Institute(IDEAS研究机构);IDEAS NCBR Sp. z o.o.(IDEAS NCBR公司)
Energy Dissipation Rate Guided Adaptive Sampling for Physics-Informed Neural Networks: Resolving Surface-Bulk Dynamics in Allen-Cahn Systems
Department of Mathematics, The Pennsylvania State University(数学系,宾夕法尼亚州立大学);Department of Mathematics, The Hong Kong University of Science and Technology(数学系,香港科学与技术大学);Department of Mathematics, University of South Carolina(数学系,南卡罗来纳大学)
Broken Ergodicity and the Violation of the Fluctuation-Dissipation Theorem Lead to Generalization Beyond Overfitting in Machine Learning
Department of Physics, University of California San Diego(加州大学圣地亚哥分校物理系);Halıcıoğlu Data Science Institute, University of California San Diego(加州大学圣地亚哥分校Halıcıoğlu数据科学研究所)