Profile

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

Stream

Mat Kelcey

Shared publicly  - 
 
We introduce Embed to Control (E2C), a method for model learning and control of non-linear dynamical systems from raw pixel images. E2C consists of a deep generative model, belonging to the family of variational autoencoders, that learns to generate image trajectories from a latent space in which the dynamics is constrained to be locally linear. Our model is derived directly from an optimal control formulation in latent space, supports long-term prediction of image sequences and exhibits strong performance on a variety of complex control problems.
Abstract: We introduce Embed to Control (E2C), a method for model learning and control of non-linear dynamical systems from raw pixel images. E2C consists of a deep generative model, belonging to the family of variational autoencoders, that learns to generate image trajectories from a latent ...
1
Add a comment...

Mat Kelcey

Shared publicly  - 
 
In this paper we study the problem of answering cloze-style questions over short documents. We introduce a new attention mechanism which uses multiplicative interactions between the query embedding and intermediate states of a recurrent neural network reader. This enables the reader to build query-specific representations of tokens in the document which are further used for answer selection. Our model, the Gated-Attention Reader, outperforms all state-of-the-art models on several large-scale benchmark datasets for this task---the CNN \& Dailymail news stories and Children's Book Test. We also provide a detailed analysis of the performance of our model and several baselines over a subset of questions manually annotated with certain linguistic features. The analysis sheds light on the strengths and weaknesses of several existing models.
Abstract: In this paper we study the problem of answering cloze-style questions over short documents. We introduce a new attention mechanism which uses multiplicative interactions between the query embedding and intermediate states of a recurrent neural network reader. This enables the reader to ...
1
Add a comment...

Mat Kelcey

Shared publicly  - 
 
Automated discovery of early visual concepts from raw image data is a major open challenge in AI research. Addressing this problem, we propose an unsupervised approach for learning disentangled representations of the underlying factors of variation. We draw inspiration from neuroscience, and show how this can be achieved in an unsupervised generative model by applying the same learning pressures as have been suggested to act in the ventral visual stream in the brain. By enforcing redundancy reduction, encouraging statistical independence, and exposure to data with transform continuities analogous to those to which human infants are exposed, we obtain a variational autoencoder (VAE) framework capable of learning disentangled factors. Our approach makes few assumptions and works well across a wide variety of datasets. Furthermore, our solution has useful emergent properties, such as zero-shot inference and an intuitive understanding of "objectness".
Abstract: Automated discovery of early visual concepts from raw image data is a major open challenge in AI research. Addressing this problem, we propose an unsupervised approach for learning disentangled representations of the underlying factors of variation. We draw inspiration from ...
1
1
Add a comment...

Mat Kelcey

Shared publicly  - 
 
Guided policy search algorithms can be used to optimize complex nonlinear policies, such as deep neural networks, without directly computing policy gradients in the high-dimensional parameter space. Instead, these methods use supervised learning to train the policy to mimic a "teacher" algorithm, such as a trajectory optimizer or a trajectory-centric reinforcement learning method. Guided policy search methods provide asymptotic local convergence guarantees by construction, but it is not clear how much the policy improves within a small, finite number of iterations. We show that guided policy search algorithms can be interpreted as an approximate variant of mirror descent, where the projection onto the constraint manifold is not exact. We derive a new guided policy search algorithm that is simpler and provides appealing improvement and convergence guarantees in simplified convex and linear settings, and show that in the more general nonlinear setting, the error in the projection step can be bounded. We provide empirical results on several simulated robotic navigation and manipulation tasks that show that our method is stable and achieves similar or better performance when compared to prior guided policy search methods, with a simpler formulation and fewer hyperparameters.
Abstract: Guided policy search algorithms can be used to optimize complex nonlinear policies, such as deep neural networks, without directly computing policy gradients in the high-dimensional parameter space. Instead, these methods use supervised learning to train the policy to mimic a "teacher" ...
1
Add a comment...

Mat Kelcey

Shared publicly  - 
 
The National Robotics Initiative (NRI) was launched 2011 and is about to celebrate its 5 year anniversary. In parallel with the NRI, the robotics community, with support from the Computing Community Consortium, engaged in a series of road mapping exercises. The first version of the roadmap appeared in September 2009; a second updated version appeared in 2013. While not directly aligned with the NRI, these road-mapping documents have provided both a useful charting of the robotics research space, as well as a metric by which to measure progress. This report sets forth a perspective of progress in robotics over the past five years, and provides a set of recommendations for the future. The NRI has in its formulation a strong emphasis on co-robot, i.e., robots that work directly with people. An obvious question is if this should continue to be the focus going forward? To try to assess what are the main trends, what has happened the last 5 years and what may be promising directions for the future a small CCC sponsored study was launched to have two workshops, one in Washington DC (March 5th, 2016) and another in San Francisco, CA (March 11th, 2016). In this report we brief summarize some of the main discussions and observations from those workshops. We will present a variety of background information in Section 2, and outline various issues related to progress over the last 5 years in Section 3. In Section 4 we will outline a number of opportunities for moving forward. Finally, we will summarize the main points in Section 5.
Donate to arXiv. One hundred percent of your contribution will fund improvements and new initiatives to benefit arXiv's global scientific community. Please join the Simons Foundation and our generous member organizations and research labs in supporting arXiv.
1
Add a comment...

Mat Kelcey

Shared publicly  - 
 
Rapid progress in machine learning and artificial intelligence (AI) has brought increasing attention to the potential impacts of AI technologies on society. In this paper we discuss one such potential impact: the problem of accidents in machine learning systems, defined as unintended and harmful behavior that may emerge from poor design of real-world AI systems. We present a list of five practical research problems related to accident risk, categorized according to whether the problem originates from having the wrong objective function ("avoiding side effects" and "avoiding reward hacking"), an objective function that is too expensive to evaluate frequently ("scalable supervision"), or undesirable behavior during the learning process ("safe exploration" and "distributional shift"). We review previous work in these areas as well as suggesting research directions with a focus on relevance to cutting-edge AI systems. Finally, we consider the high-level question of how to think most productively about the safety of forward-looking applications of AI.
Abstract: Rapid progress in machine learning and artificial intelligence (AI) has brought increasing attention to the potential impacts of AI technologies on society. In this paper we discuss one such potential impact: the problem of accidents in machine learning systems, defined as unintended ...
1
Add a comment...
In their circles
288 people
Have them in circles
366 people
Chris Balogh's profile photo
Mike Williams's profile photo
Hyunsik Choi's profile photo
Richard Cole's profile photo
Noun Choeun's profile photo
Simon Pagel's profile photo
Mirtha Estela Munayco Perez's profile photo
L. Garcia Dennis's profile photo
Victor Huckabee's profile photo

Mat Kelcey

Shared publicly  - 
 
Recent work has shown that deep neural networks are capable of approximating both value functions and policies in reinforcement learning domains featuring continuous state and action spaces. However, to the best of our knowledge no previous work has succeeded at using deep neural networks in structured (parameterized) continuous action spaces. To fill this gap, this paper focuses on learning within the domain of simulated RoboCup soccer, which features a small set of discrete action types, each of which is parameterized with continuous variables. The best learned agent can score goals more reliably than the 2012 RoboCup champion agent. As such, this paper represents a successful extension of deep reinforcement learning to the class of parameterized action space MDPs.

Abstract: Recent work has shown that deep neural networks are capable of approximating both value functions and policies in reinforcement learning domains featuring continuous state and action spaces. However, to the best of our knowledge no previous work has succeeded at using deep neural ...
1
Add a comment...

Mat Kelcey

Shared publicly  - 
 
We present an approach to training neural networks to generate sequences using actor-critic methods from reinforcement learning (RL). Current log-likelihood training methods are limited by the discrepancy between their training and testing modes, as models must generate tokens conditioned on their previous guesses rather than the ground-truth tokens. We address this problem by introducing a \textit{critic} network that is trained to predict the value of an output token, given the policy of an \textit{actor} network. This results in a training procedure that is much closer to the test phase, and allows us to directly optimize for a task-specific score such as BLEU. Crucially, since we leverage these techniques in the supervised learning setting rather than the traditional RL setting, we condition the critic network on the ground-truth output. We show that our method leads to improved performance on both a synthetic task, and for German-English machine translation. Our analysis paves the way for such methods to be applied in natural language generation tasks, such as machine translation, caption generation, and dialogue modelling.
Abstract: We present an approach to training neural networks to generate sequences using actor-critic methods from reinforcement learning (RL). Current log-likelihood training methods are limited by the discrepancy between their training and testing modes, as models must generate tokens ...
1
Add a comment...

Mat Kelcey

Shared publicly  - 
 
Variational Autoencoders are powerful models for unsupervised learning. However deep models with several layers of dependent stochastic variables are difficult to train which limits the improvements obtained using these highly expressive models. We propose a new inference model, the Ladder Variational Autoencoder, that recursively corrects the generative distribution by a data dependent approximate likelihood in a process resembling the recently proposed Ladder Network. We show that this model provides state of the art predictive log-likelihood and tighter log-likelihood lower bound compared to the purely bottom-up inference in layered Variational Autoencoders and other generative models. We provide a detailed analysis of the learned hierarchical latent representation and show that our new inference model is qualitatively different and utilizes a deeper more distributed hierarchy of latent variables. Finally, we observe that batch normalization and deterministic warm-up (gradually turning on the KL-term) are crucial for training variational models with many stochastic layers.

Abstract: Variational Autoencoders are powerful models for unsupervised learning. However deep models with several layers of dependent stochastic variables are difficult to train which limits the improvements obtained using these highly expressive models. We propose a new inference model, ...
1
Add a comment...

Mat Kelcey

Shared publicly  - 
 
Training state-of-the-art, deep neural networks is computationally expensive. One way to reduce the training time is to normalize the activities of the neurons. A recently introduced technique called batch normalization uses the distribution of the summed input to a neuron over a mini-batch of training cases to compute a mean and variance which are then used to normalize the summed input to that neuron on each training case. This significantly reduces the training time in feed-forward neural networks. However, the effect of batch normalization is dependent on the mini-batch size and it is not obvious how to apply it to recurrent neural networks. In this paper, we transpose batch normalization into layer normalization by computing the mean and variance used for normalization from all of the summed inputs to the neurons in a layer on a single training case. Like batch normalization, we also give each neuron its own adaptive bias and gain which are applied after the normalization but before the non-linearity. Unlike batch normalization, layer normalization performs exactly the same computation at training and test times. It is also straightforward to apply to recurrent neural networks by computing the normalization statistics separately at each time step. Layer normalization is very effective at stabilizing the hidden state dynamics in recurrent networks. Empirically, we show that layer normalization can substantially reduce the training time compared with previously published techniques.
Abstract: Training state-of-the-art, deep neural networks is computationally expensive. One way to reduce the training time is to normalize the activities of the neurons. A recently introduced technique called batch normalization uses the distribution of the summed input to a neuron over a ...
2
Add a comment...

Mat Kelcey

Shared publicly  - 
 
In just three years, Variational Autoencoders (VAEs) have emerged as one of the most popular approaches to unsupervised learning of complicated distributions. VAEs are appealing because they are built on top of standard function approximators (neural networks), and can be trained with stochastic gradient descent. VAEs have already shown promise in generating many kinds of complicated data, including handwritten digits, faces, house numbers, CIFAR images, physical models of scenes, segmentation, and predicting the future from static images. This tutorial introduces the intuitions behind VAEs, explains the mathematics behind them, and describes some empirical behavior. No prior knowledge of variational Bayesian methods is assumed.
Abstract: In just three years, Variational Autoencoders (VAEs) have emerged as one of the most popular approaches to unsupervised learning of complicated distributions. VAEs are appealing because they are built on top of standard function approximators (neural networks), and can be trained with ...
1
Add a comment...

Mat Kelcey

Shared publicly  - 
 
Deep reinforcement learning has been shown to be a powerful framework for learning policies from complex high-dimensional sensory inputs to actions in complex tasks, such as the Atari domain. In this paper, we explore output representation modeling in the form of temporal abstraction to improve convergence and reliability of deep reinforcement learning approaches. We concentrate on macro-actions, and evaluate these on different Atari 2600 games, where we show that they yield significant improvements in learning speed. Additionally, we show that they can even achieve better scores than DQN. We offer analysis and explanation for both convergence and final results, revealing a problem deep RL approaches have with sparse reward signals.
Abstract: Deep reinforcement learning has been shown to be a powerful framework for learning policies from complex high-dimensional sensory inputs to actions in complex tasks, such as the Atari domain. In this paper, we explore output representation modeling in the form of temporal abstraction ...
1
Add a comment...
Mat's Collections
People
In their circles
288 people
Have them in circles
366 people
Chris Balogh's profile photo
Mike Williams's profile photo
Hyunsik Choi's profile photo
Richard Cole's profile photo
Noun Choeun's profile photo
Simon Pagel's profile photo
Mirtha Estela Munayco Perez's profile photo
L. Garcia Dennis's profile photo
Victor Huckabee'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 Google Brain on robotics learning. Previously I work on knowledge extraction & question answer systems.
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