• Publications
  • Influence
Neural Audio Synthesis of Musical Notes with WaveNet Autoencoders
TLDR
A powerful new WaveNet-style autoencoder model is detailed that conditions an autoregressive decoder on temporal codes learned from the raw audio waveform, and NSynth, a large-scale and high-quality dataset of musical notes that is an order of magnitude larger than comparable public datasets is introduced.
Pommerman: A Multi-Agent Playground
TLDR
Pommerman, a multi-agent environment based on the classic console game Bomberman, consists of a set of scenarios, each having at least four players and containing both cooperative and competitive aspects.
Backplay: "Man muss immer umkehren"
TLDR
The approach, Backplay, uses a single demonstration to construct a curriculum for a given task, and analytically characterize the types of environments where Backplay can improve training speed and compare favorably to other competitive methods known to improve sample efficiency.
Capacity, Bandwidth, and Compositionality in Emergent Language Learning
TLDR
The hypothesis is that there should be a specific range of model capacity and channel bandwidth that induces compositional structure in the resulting language and consequently encourages systematic generalization.
Audio Deepdream: Optimizing raw audio with convolutional networks
TLDR
This work has followed in the footsteps of Van den Oord et al and trained a network to predict embeddings that were themselves the result of a collaborative filtering model, which creates a chain of differentiable functions from raw audio to high level features.
Compositionality and Capacity in Emergent Languages
TLDR
This paper investigates the learning biases that affect the efficacy and compositionality in multi-agent communication in addition to the communicative bandwidth and explores how the capacity of a neural network impacts its ability to learn a compositional language.
Probing the State of the Art: A Critical Look at Visual Representation Evaluation
TLDR
This work shows that this test is insufficient and that models which perform poorly on linear classification can perform strongly (weakly) on more involved tasks like temporal activity localization.
Vehicle Community Strategies
TLDR
This work considers self-driving cars coordinating with each other and focuses on how communication influences the agents' collective behavior, finding that communication helps (most) with adverse conditions.
In-Distribution Interpretability for Challenging Modalities
TLDR
This work displays the flexibility of the intuitive framework which utilizes generative models to improve on the meaningfulness of such explanations of deep neural networks: music and physical simulations of urban environments.
Vehicle Communication Strategies for Simulated Highway Driving
TLDR
This work considers self-driving cars coordinating with each other and focuses on how communication influences the agents' collective behavior, finding that communication helps (most) with adverse conditions.
...
1
2
...