• Publications
  • Influence
Reducing the variance in online optimization by transporting past gradients
The idea of implicit gradient transport (IGT) which transforms gradients computed at previous iterates into gradients evaluated at the current iterate without using the Hessian explicitly is proposed which yields the optimal asymptotic convergence rate for online stochastic optimization in the restricted setting where the Hessians of all component functions are equal. Expand
An Introduction to Distributed Deep Learning
This blog post introduces the fundamentals of distributed deep learning and presents some real-world applications. With the democratisation of deep learning methods in the last decade, large andExpand
Embedding Adaptation is Still Needed for Few-Shot Learning
This work proposes ATG, a principled clustering method to defining train and test tasksets without additional human knowledge, and empirically demonstrates the effectiveness of ATG in generating tasksets that are easier, in-between, or harder than existing benchmarks, including those that rely on semantic information. Expand
learn2learn: A Library for Meta-Learning Research
Meta-learning researchers face two fundamental issues in their empirical work: prototyping and reproducibility. Researchers are prone to make mistakes when prototyping new algorithms and tasksExpand
When MAML Can Adapt Fast and How to Assist When It Cannot
This work finds MAML adapts better with a deep architecture even if the tasks need only a shallow one (and thus, no representation learning is needed), and also finds that upper layers enable fast adaptation by being meta-learned to perform adaptive gradient update when generalizing to new tasks. Expand
Decoupling Adaptation from Modeling with Meta-Optimizers for Meta Learning
This work begins with an experimental analysis of MAML, finding that deep models are crucial for its success, even given sets of simple tasks where a linear model would suffice on any individual task. Expand
Analyzing the Variance of Policy Gradient Estimators for the Linear-Quadratic Regulator
This work studies the variance of the REINFORCE policy gradient estimator in environments with continuous state and action spaces, linear dynamics, quadratic cost, and Gaussian noise to derive bounds on the estimator variance in terms of the environment and noise parameters. Expand
Uniform Sampling over Episode Difficulty
This paper proposes a method to approximate episode sampling distributions based on their difficulty and finds that sampling uniformly over episode difficulty outperforms other sampling schemes, including curriculum and easy-/hard-mining. Expand
Shapechanger: Environments for Transfer Learning
We present Shapechanger, a library for transfer reinforcement learning specifically designed for robotic tasks. We consider three types of knowledge transfer---from simulation to simulation, fromExpand
Accelerating SGD for Distributed Deep-Learning Using Approximated Hessian Matrix
This work suggests that novel strategies for combining gradients provide further information on the loss surface and underline advantages and challenges of second-order methods for large stochastic optimization problems. Expand