Profile

Cover photo
Mat Kelcey
Works at Google
Lives in San Francisco Bay Area
365 followers
AboutPostsCollections

Stream

Mat Kelcey

Shared publicly  - 
 
general placement of the sonars while i'm waiting for the 3d printed mounting to arrive. the control pcb is just floating, i should bolt it down...
1
James Brooks's profile photoMat Kelcey's profile photo
2 comments
 
yeah i been wondering about it. being able to get stuff printed online for reasonable prices makes it hard to justify until i start doing a fair bit. we have a set at work but they keep breaking down all the time
Add a comment...

Mat Kelcey

Shared publicly  - 
 
soldering done!
1
Add a comment...

Mat Kelcey

Shared publicly  - 
 
"Dueling Network Architectures for Deep Reinforcement Learning"

In recent years there have been many successes of using deep representations in reinforcement learning. Still, many of these applications use conventional architectures, such as convolutional networks, LSTMs, or auto-encoders. In this paper, we present a new neural network architecture for model-free reinforcement learning. Our dueling network represents two separate estimators: one for the state value function and one for the state-dependent action advantage function. The main benefit of this factoring is to generalize learning across actions without imposing any change to the underlying reinforcement learning algorithm. Our results show that this architecture leads to better policy evaluation in the presence of many similar-valued actions. Moreover, the dueling architecture enables our RL agent to outperform the state-of-the-art on the Atari 2600 domain.

Abstract: In recent years there have been many successes of using deep representations in reinforcement learning. Still, many of these applications use conventional architectures, such as convolutional networks, LSTMs, or auto-encoders. In this paper, we present a new neural network architecture ...
1
Add a comment...

Mat Kelcey

Shared publicly  - 
 
In this paper we propose a new method for regularizing autoencoders by imposing an arbitrary prior on the latent representation of the autoencoder. Our method, named "adversarial autoencoder", uses the recently proposed generative adversarial networks (GAN) in order to match the aggregated posterior of the hidden code vector of the autoencoder with an arbitrary prior. Matching the aggregated posterior to the prior ensures that there are no "holes" in the prior, and generating from any part of prior space results in meaningful samples. As a result, the decoder of the adversarial autoencoder learns a deep generative model that maps the imposed prior to the data distribution. We show how adversarial autoencoders can be used to disentangle style and content of images and achieve competitive generative performance on MNIST, Street View House Numbers and Toronto Face datasets.
Abstract: In this paper we propose a new method for regularizing autoencoders by imposing an arbitrary prior on the latent representation of the autoencoder. Our method, named "adversarial autoencoder", uses the recently proposed generative adversarial networks (GAN) in order to match the ...
1
Add a comment...

Mat Kelcey

Shared publicly  - 
 
new PCB from http://fab.fritzing.org/ done! will handle the 3 HC-SR04 sonars
1
Add a comment...

Mat Kelcey

Shared publicly  - 
 
"Deep Reinforcement Learning in Large Discrete Action Spaces"

Being able to reason in an environment with a large number of discrete actions is essential to bringing reinforcement learning to a larger class of problems. Recommender systems, industrial plants and language models are only some of the many real world tasks involving large numbers of discrete actions for which current methods are difficult or even often impossible to apply. An ability to generalize over the set of actions as well as sub-linear complexity relative to the size of the set are both necessary to handle such tasks. Current approaches are not able to provide both of these, which motivates the work in this paper. Our proposed approach leverages prior information about the actions to embed them in a continuous space upon which it can generalize. Additionally, approximate nearest-neighbor methods allow for logarithmic-time lookup complexity relative to the number of actions, which is necessary for time-wise tractable training. This combined approach allows reinforcement learning methods to be applied to large-scale learning problems previously intractable with current methods. We demonstrate our algorithm's abilities on a series of tasks having up to one million actions.
Abstract: Being able to reason in an environment with a large number of discrete actions is essential to bringing reinforcement learning to a larger class of problems. Recommender systems, industrial plants and language models are only some of the many real-world tasks involving large numbers of ...
1
Add a comment...
Have them in circles
365 people
Darren D'Rozario's profile photo
tracy max's profile photo
Evan Bottcher's profile photo
Kari Johnson's profile photo
Matt Siegel's profile photo
Devendra Singh Sachan's profile photo
Regina Ruby  Bernice Harris surprise's profile photo
Shao Hong Peh's profile photo
Mark Mansour's profile photo

Mat Kelcey

Shared publicly  - 
 
"Hierarchical Memory Networks"
Memory networks are neural networks with an explicit memory component that can be both read and written to by the network. The memory is often addressed in a soft way using a softmax function, making end-to-end training with backpropagation possible. However, this is not computationally scalable for applications which require the network to read from extremely large memories. On the other hand, it is well known that hard attention mechanisms based on reinforcement learning are challenging to train successfully. In this paper, we explore a form of hierarchical memory network, which can be considered as a hybrid between hard and soft attention memory networks. The memory is organized in a hierarchical structure such that reading from it is done with less computation than soft attention over a flat memory, while also being easier to train than hard attention over a flat memory. Specifically, we propose to incorporate Maximum Inner Product Search (MIPS) in the training and inference procedures for our hierarchical memory network. We explore the use of various state-of-the art approximate MIPS techniques and report results on SimpleQuestions, a challenging large scale factoid question answering task.
Abstract: Memory networks are neural networks with an explicit memory component that can be both read and written to by the network. The memory is often addressed in a soft way using a softmax function, making end-to-end training with backpropagation possible.
2
Add a comment...

Mat Kelcey

Shared publicly  - 
 
Despite recent breakthroughs in the applications
of deep neural networks, one setting that presents
a persistent challenge is that of “one-shot learning.”
Traditional gradient-based networks require
a lot of data to learn, often through extensive iterative
training. When new data is encountered,
the models must inefficiently relearn their parameters
to adequately incorporate the new information
without catastrophic interference. Architectures
with augmented memory capacities, such as
Neural Turing Machines (NTMs), offer the ability
to quickly encode and retrieve new information,
and hence can potentially obviate the downsides
of conventional models. Here, we demonstrate
the ability of a memory-augmented neural
network to rapidly assimilate new data, and
leverage this data to make accurate predictions
after only a few samples. We also introduce a
new method for accessing an external memory
that focuses on memory content, unlike previous
methods that additionally use memory locationbased
focusing mechanisms.
Abstract: Despite recent breakthroughs in the applications of deep neural networks, one setting that presents a persistent challenge is that of "one-shot learning." Traditional gradient-based networks require a lot of data to learn, often through extensive iterative training.
1
Add a comment...

Mat Kelcey

Shared publicly  - 
 
Reinforcement Learning: A Survey


1
Add a comment...

Mat Kelcey

Shared publicly  - 
 
( the setup my PCB is replacing, too much going on with that breadboard!!! )
1
Add a comment...

Mat Kelcey

Shared publicly  - 
 
A C++ ROS node wrapper for publishing sonar readings from a set of sonars (one per topic)
ros-hc-sr04-node - raspberry pi c++ ROS hc-sr04 node
1
Add a comment...
Mat's Collections
People
Have them in circles
365 people
Darren D'Rozario's profile photo
tracy max's profile photo
Evan Bottcher's profile photo
Kari Johnson's profile photo
Matt Siegel's profile photo
Devendra Singh Sachan's profile photo
Regina Ruby  Bernice Harris surprise's profile photo
Shao Hong Peh's profile photo
Mark Mansour's profile photo
Work
Occupation
Data nerd wannabe
Skills
Machine learning, natural language processing, information retrieval, distributed systems.
Employment
  • Google
    Software Engineer, present
  • Wavii
    Software Engineer
  • Amazon Web Services
    Software Engineer
  • Lonely Planet
    Software Engineer
  • Sensis
    Software Engineer
  • Distra
    Software Engineer
  • Nokia
    Software Engineer
  • Australian Stock Exchange
    Software Engineer
Basic Information
Gender
Decline to State
Story
Introduction
I work in the Machine Intelligence group at Google building as-large-as-I-can-get neural networks for knowledge extraction.
Places
Map of the places this user has livedMap of the places this user has livedMap of the places this user has lived
Currently
San Francisco Bay Area
Previously
seattle - melbourne - calgary - london - sydney - hobart
Links
Contributor to