arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1904.03081cs.LGcs.CVstat.ML
收藏

基于能量耗散的神经网络控制

Controlling Neural Networks via Energy Dissipation

University of Siegen(锡根大学);TU Munich(慕尼黑工业大学)

Michael Moeller, Thomas Möllenhoff, Daniel Cremers

摘要:

在过去的十年中,在深度学习技术的帮助下,在解决各种计算机视觉问题方面取得了巨大的成功。最近,许多工作已经证明,基于学习的方法与合适的网络架构,甚至表现出优越的性能(不适定)图像重建问题的解决方案,如去模糊,超分辨率,或医学图像重建。然而,纯粹基于学习的方法的缺点是,它们不能为训练的网络提供可证明的保证,以在推理过程中遵循给定的数据形成过程。在这项工作中,我们提出了能量耗散网络,迭代计算下降方向相对于一个给定的成本函数或能量在当前估计的重建。因此,自适应步长规则(诸如线搜索)连同合适的迭代次数可以保证重建遵循以任意精度在能量中编码的给定数据形成模型,并且因此即使在测试时间期间也控制模型的行为。我们证明了在标准假设下,使用网络预测的方向下降收敛(线性)到能量的全局最小值。我们说明了所提出的方法的有效性,在实验中对单图像超分辨率和计算机断层扫描(CT)重建,并进一步说明凸可行性问题的扩展。

英文摘要:

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.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑