• Publications
  • Influence
TADAM: Task dependent adaptive metric for improved few-shot learning
TLDR
This work identifies that metric scaling and metric task conditioning are important to improve the performance of few-shot algorithms and proposes and empirically test a practical end-to-end optimization procedure based on auxiliary task co-training to learn a task-dependent metric space.
Neural Autoregressive Flows
TLDR
It is demonstrated that the proposed neural autoregressive flows (NAF) are universal approximators for continuous probability distributions, and their greater expressivity allows them to better capture multimodal target distributions.
PAC-Bayesian Theory Meets Bayesian Inference
TLDR
For the negative log-likelihood loss function, it is shown that the minimization of PAC-Bayesian generalization risk bounds maximizes the Bayesian marginal likelihood.
Learning Heuristics for the TSP by Policy Gradient
TLDR
The neural combinatorial optimization framework is extended to solve the traveling salesman problem (TSP) and the performance of the proposed framework alone is generally as good as high performance heuristics (OR-Tools).
WikiReading: A Novel Large-scale Language Understanding Task over Wikipedia
TLDR
This work presents WIKIREADING, a large-scale natural language understanding task and publicly-available dataset with 18 million instances, and compares various state-of-the-art DNNbased architectures for document classification, information extraction, and question answering.
Quantifying the Carbon Emissions of Machine Learning
TLDR
This work presents their Machine Learning Emissions Calculator, a tool for the community to better understand the environmental impact of training ML models and concrete actions that individual practitioners and organizations can take to mitigate their carbon emissions.
Coarse-to-Fine Question Answering for Long Documents
TLDR
A framework for question answering that can efficiently scale to longer documents while maintaining or even improving performance of state-of-the-art models is presented and sentence selection is treated as a latent variable trained jointly from the answer only using reinforcement learning.
Tackling Climate Change with Machine Learning
TLDR
From smart grids to disaster management, high impact problems where existing gaps can be filled by ML are identified, in collaboration with other fields, to join the global effort against climate change.
Embedding Propagation: Smoother Manifold for Few-Shot Classification
TLDR
This work empirically shows that embedding propagation yields a smoother embedding manifold, and shows that applying embedding propagate to a transductive classifier achieves new state-of-the-art results in mini-Imagenet, tiered-Imageet, Imagenet-FS, and CUB.
Differentiable Causal Discovery from Interventional Data
TLDR
This work proposes a neural network-based method for discovering causal relationships in data that can leverage interventional data and illustrates the flexibility of the continuous-constrained framework by taking advantage of expressive neural architectures such as normalizing flows.
...
...