Machine Learning + X Seminars

Machine Learning + X seminars are held byweekly.

Selected presentations from past seminars are posted blow. Yeonjong Shin

  • Quantifying the generalization error in deep learning in terms of data distribution and neural network smoothness. Pengzhan Jin
  • MoGlow: Probabilistic and controllable motion synthesis using normalising flows.
  • Super-resolution reconstruction of turbulent flows with machine learning.
  • Fast and Scalable Bayesian Deep Learning by Weight-Perturbation in Adam.
  • Bridging Finite Element and Machine Learning Modeling: Stress Prediction of Arterial Walls in Atherosclerosis
  • Deep relaxation: partial differential equations for optimizing deep neural networks.
  • Data driven nonlinear dynamical systems identification using multi-step CLDNN.
  • Recent applications of neural networks in biological systems
  • A synthetic turbulent inflow generator using machine learning
  • Spectral Normalization for Generative Adversarial Networks
  • Generative Modeling using the Sliced Wasserstein Distance.
  • Sliced Wasserstein Generative Models.
  • Understanding training and generalization in deep learning by Fourier analysis.
  • An analysis of training and generalization errors in shallow and deep networks.
  • Deep vs. Shallow Networks: an Approximation Theory Perspective.
  • Predicting the solutions of heterogeneous elliptic PDEs with a confidence interval by probabilistic convolutional neural networks.
  •  Error bounds for approximations with deep ReLU neural networks in W^{s,p} norms.
  • Universal Function Approximation by Deep Neural Nets with Bounded Width and ReLU Activations .
  • Assessment of End-to-End and Sequential Data-driven Learning of Fluid Flows.
  • Boost your research through SciCoNet and SumsJob.  Lu Lu
  • An inverse problem framework for extracting phonon properties from thermal spectroscopy measurments .  Mojtaba Forghani
  • A review of definitions of fractional derivatives and other operators.  Ehsan Kharazmi
  • Paper Review: Optimal approximation of continuous functions by very deep ReLU networks.  Dmitry Yarotsky 
  • Mark Thachuk

    Department of Chemistry, University of British Columbia, Canada

    Title: Coarse Graining Independent Particles using Mori-Zwanzig Theory

  • Ameya Jagtap

    Paper Review: 
    Artificial Neural Networks Trained Through Deep Reinforcement Learning Discover Control Strategies for Active Flow Control (arXiv:1808.07664v5)

  • Yeonjong Shin

    Paper Review: Elimination of all bad local minima in deep learning by Kenji Kawaguchi and Leslie Pack Kaelbling

  • Mojtaba Forghani

    Mechanical Engineering Department, MIT

    Title: An inverse problem framework for extracting phonon properties from thermal spectroscopy measurements.

  • Drew Linsley, Junkyung Kim & Thomas Serre

    TITLE: Recurrent neural networks for visual processing

  • Xiang Li

    TITLE: The Acquisition and Uncertainty Quantification of Land Surface Evapotranspiration at the Satellite Pixel Scale

  • Prof. Yosuke Hasegawa

    TITLE: Vascular Network of Zebrafish Brain

  • Ludger Paehler, Technische Universität München
    TITLE: Multifidelity and Machine Learning for Turbulent Flows
  • Alireza Yazdani

    TITLE: Data-Driven Multiscale Modeling in Physical and Biological Systems

  • Dr. Yu-Hang Tang, 
    Lawrence Berkeley National Laboratory
    TITLE: 
    Prediction of atomization energy using graph kernel and active learning
  • Guofei Pang -  Paper Review:  Adversarial Examples, Uncertainty, and Transfer Testing Robustness in Gaussian Process Hybrid Deep Networks by John Bradshaw, Alexander G. de G. Matthews, Zoubin Ghahramani
  • Guofei Pang - Image recognition: Defense adversarial attacks using Generative Adversarial Network (GAN)
  • Lu Lu - 4 Years of Generative Adversarial Networks (GANs)
  • Yanhui Su - Approximation theory in neural networks
  • Zhiping Mao - Learning networks of stochastic differential equations
  • Guofei Pang - When a neural net meets a Gaussian process
Related Files
application/pdf icon4 Years of Generative Adversarial Networks (GANs)
application/vnd.openxmlformats-officedocument.presentationml.presentation icon Image recognition: Defense adversarial attacks using Generative Adversarial Network (GAN)
application/pdf iconLearning Networks of Stochastic Differential Equations.pdf
application/pdf iconApproximation theory in neural networks.pdf
application/pdf iconWhen a neural net meets a Gaussian process