Object-sensitive Deep Reinforcement Learning

@inproceedings{Li2017ObjectsensitiveDR,
  title={Object-sensitive Deep Reinforcement Learning},
  author={Yuezhang Li and Katia P. Sycara and Rahul Radhakrishnan Iyer},
  booktitle={GCAI},
  year={2017}
}
Deep reinforcement learning has become popular over recent years, showing superiority on different visual-input tasks such as playing Atari games and robot navigation. [] Key Method This approach can be adapted to any existing deep reinforcement learning frameworks. State-of-the-art results are shown in experiments on Atari games. We also propose a new approach called "object saliency maps" to visually explain the actions made by deep reinforcement learning agents.

Figures and Tables from this paper

Self-Supervised Object-Level Deep Reinforcement Learning
TLDR
A probabilistic object-based model of environments is proposed and human object priors are used to develop an efficient self-supervised algorithm for maximum likelihood estimation of the model parameters from observations and for inferring objects directly from the perceptual stream.
Unsupervised Video Object Segmentation for Deep Reinforcement Learning
TLDR
A new technique for deep reinforcement learning that automatically detects moving objects and uses the relevant information for action selection and is more interpretable than policies that directly map images to actions or values with a black box neural network.
Verbal Explanations for Deep Reinforcement Learning Neural Networks with Attention on Extracted Features
TLDR
This paper focuses on generating verbal (natural language) descriptions and explanations of deep reinforcement learning policies, and shows that the proposed encoder outperforms two baseline encoders on the capability of distinguishing the game state images.
Towards Better Interpretability in Deep Q-Networks
TLDR
An interpretable neural network architecture for Q- learning is proposed which provides a global explanation of the model's behavior using key-value memories, attention and reconstructible embeddings and can reach training rewards comparable to the state-of-the-art deep Q-learning models.
Cognitive Architecture for Video Games
TLDR
A novel architecture for FORL that understands the content of raw frames and can recognize both Goomba and Mario and classify them using incremental training is proposed, developed in the Super Mario Brothers video game.
Transparency and Explanation in Deep Reinforcement Learning Neural Networks
TLDR
A novel and general method that incorporates explicit object recognition processing into deep reinforcement learning models, forms the basis for the development of "object saliency maps", to provide visualization of internal states of DRLNs, thus enabling the formation of explanations and can be incorporated in any existing deep reinforcementlearning framework is proposed.
Explainable robotic systems: Interpreting outcome-focused actions in a reinforcement learning scenario
TLDR
This work focuses on the decision-making process of a reinforcement learning agent performing a navigation task in a robotic scenario, and uses the probability of success computed by three different proposed approaches: memory-based, learning- based, and phenomenological-based.
Verifiably safe exploration for end-to-end reinforcement learning
TLDR
A first approach toward enforcing formal safety constraints on end-to-end policies with visual inputs is contributed, drawing on recent advances in object detection and automated reasoning for hybrid dynamical systems.
Task-Relevant Object Discovery and Categorization for Playing First-person Shooter Games
TLDR
This paper presents a new and efficient method for learning low-dimensional representations of first-person shooter video games using raw screen images as observations and keyboard inputs as actions using the deep Q-network (DQN).
...
1
2
3
...

References

SHOWING 1-10 OF 33 REFERENCES
Playing FPS Games with Deep Reinforcement Learning
TLDR
This paper presents the first architecture to tackle 3D environments in first-person shooter games, that involve partially observable states, and substantially outperforms built-in AI agents of the game as well as humans in deathmatch scenarios.
Asynchronous Methods for Deep Reinforcement Learning
TLDR
A conceptually simple and lightweight framework for deep reinforcement learning that uses asynchronous gradient descent for optimization of deep neural network controllers and shows that asynchronous actor-critic succeeds on a wide variety of continuous motor control problems as well as on a new task of navigating random 3D mazes using a visual input.
Dueling Network Architectures for Deep Reinforcement Learning
TLDR
This paper presents a new neural network architecture for model-free reinforcement learning that leads to better policy evaluation in the presence of many similar-valued actions and enables the RL agent to outperform the state-of-the-art on the Atari 2600 domain.
Human-level control through deep reinforcement learning
TLDR
This work bridges the divide between high-dimensional sensory inputs and actions, resulting in the first artificial agent that is capable of learning to excel at a diverse array of challenging tasks.
Deep Reinforcement Learning with Double Q-Learning
TLDR
This paper proposes a specific adaptation to the DQN algorithm and shows that the resulting algorithm not only reduces the observed overestimations, as hypothesized, but that this also leads to much better performance on several games.
1 Supervised Actor-Critic Reinforcement Learning
Editor’s Summary: Chapter ?? introduced policy gradients as a way to improve on stochastic search of the policy space when learning. This chapter presents supervised actor-critic reinforcement
On learning to localize objects with minimal supervision
TLDR
This paper proposes a new method that achieves this goal with only image-level labels of whether the objects are present or not, and combines a discriminative submodular cover problem for automatically discovering a set of positive object windows with a smoothed latent SVM formulation.
Generalization in Reinforcement Learning: Successful Examples Using Sparse Coarse Coding
TLDR
It is concluded that reinforcement learning can work robustly in conjunction with function approximators, and that there is little justification at present for avoiding the case of general λ.
Visualizing Higher-Layer Features of a Deep Network
TLDR
This paper contrast and compare several techniques applied on Stacked Denoising Autoencoders and Deep Belief Networks, trained on several vision datasets, and shows that good qualitative interpretations of high level features represented by such models are possible at the unit level.
Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation
TLDR
This paper proposes a simple and scalable detection algorithm that improves mean average precision (mAP) by more than 30% relative to the previous best result on VOC 2012 -- achieving a mAP of 53.3%.
...
1
2
3
4
...