Talks

  1. Riashat Islam, Peter Henderson, Maziar Gomrokchi, Doina Precup “Reproducibility of Policy Gradient Methods in Reinforcement Learning”. Contributed talk at ICML 2017 Reinforcement Learning workshop (Slides)
  2. Riashat Islam, Peter Henderson. “Practical Tutorial on Using Reinforcement Learning Algorithms for Continuous Control”. Reinforcement Learning Summer School (RLSS), Montreal, Canada, 2017 (Slides, Code)
  3. Riashat Islam. “Predictive Knowledge and Abstraction in Reinforcement Learning”. MPhil MLSALT Reading Group. University of Cambridge, 2016 (Slides)
  4. Riashat Islam. “Active Learning with Bayesian Convolutional Neural Networks using Uncertainty in Deep Learning”. University of Cambridge, UK 2016
  5. Riashat Islam. “Preordering for Statistical Machine Translation”. University of Cambridge, UK, 2016
  6. Riashat Islam. “Bayesian Active Learning in Deep Learning”. Independent University Bangladesh (IUB), Bangladesh, 2016
  7. Riashat Islam. “Improving Convergence of Deterministic Policy Gradients in Reinforcement Learning”. University College London, UK, 2015
  8. Riashat Islam. “Improved State Estimation of a Resilient Spacecraft Executive”. SURF, California Institute of Technology, USA, 2015
  9. Riashat Islam. “A Unifying Review of Linear Gaussian Models”. MPhil MLSALT, University of Cambridge, UK, 2015
  10. Riashat Islam. “Cost Sensitive Decision Tree Classification for Sepsis Data”. SRE, Johns Hopkins University, USA, 2014