Search Results for author:
Found 28 papers, 5 papers with code
Learning to Accelerate by the Methods of Step-size Planning
In the second part of this paper, we propose a new class of methods of accelerating gradient descent that have some distinctiveness from existing techniques.
Explainable artificial intelligence for autonomous driving: An overview and guide for future research directions
First, we provide a thorough overview of the state-of-the-art studies on XAI for autonomous driving.
Towards Safe, Explainable, and Regulated Autonomous Driving
There has been growing interest in the development and deployment of autonomous vehicles on roads over the last few years, encouraged by the empirical successes of powerful artificial intelligence techniques (AI), especially in the applications of deep learning and reinforcement learning.
Exploring the Robustness of Distributional Reinforcement Learning against Noisy State Observations
In real scenarios, state observations that an agent observes may contain measurement errors or adversarial noises, misleading the agent to take suboptimal actions or even collapse while training.
Distributional Reinforcement Learning
reinforcement-learning
Exploring the Training Robustness of Distributional Reinforcement Learning against Noisy State Observations
In real scenarios, state observations that an agent observes may contain measurement errors or adversarial noises, misleading the agent to take suboptimal actions or even collapse while training.
Breaking the Deadly Triad with a Target Network
The deadly triad refers to the instability of a reinforcement learning algorithm when it employs off-policy learning, function approximation, and bootstrapping simultaneously.
Measuring and mitigating interference in reinforcement learning
Catastrophic interference is common in many network-based learning systems, and many proposals exist for mitigating it.
Beyond Prioritized Replay: Sampling States in Model-Based RL via Simulated Priorities
The prioritized Experience Replay (ER) method has attracted great attention; however, there is little theoretical understanding of such prioritization strategy and why they help.
Variance-Reduced Off-Policy Memory-Efficient Policy Search
To achieve variance-reduced off-policy-stable policy optimization, we propose an algorithm family that is memory-efficient, stochastically variance-reduced, and capable of learning from off-policy samples.
Beyond Prioritized Replay: Sampling States in Model-Based Reinforcement Learning via Simulated Priorities
The prioritized Experience Replay (ER) method has attracted great attention; however, there is little theoretical understanding about why it can help and its limitations.
Towards a practical measure of interference for reinforcement learning
In this work, we provide a definition of interference for control in reinforcement learning.
Weakly Supervised Few-shot Object Segmentation using Co-Attention with Visual and Semantic Embeddings
Our results show that few-shot segmentation benefits from utilizing word embeddings, and that we are able to perform few-shot segmentation using stacked joint visual semantic processing with weak image-level labels.
One-Shot Weakly Supervised Video Object Segmentation
Conventional few-shot object segmentation methods learn object segmentation from a few labelled support images with strongly labelled segmentation masks.
Provably Convergent Two-Timescale Off-Policy Actor-Critic with Function Approximation
With the help of the emphasis critic and the canonical value function critic, we show convergence for COF-PAC, where the critics are linear and the actor can be nonlinear.
Mapless Navigation among Dynamics with Social-safety-awareness: a reinforcement learning approach from 2D laser scans
We observe that our method demonstrates time-efficient path planning behavior with high success rate in mapless navigation tasks.
Robotics
Discounted Reinforcement Learning Is Not an Optimization Problem
Discounted reinforcement learning is fundamentally incompatible with function approximation for control in continuing tasks.
Is Fast Adaptation All You Need?
Gradient-based meta-learning has proven to be highly effective at learning model initializations, representations, and update rules that allow fast adaptation from a few samples.
Three-Head Neural Network Architecture for AlphaZero Learning
A three-head network architecture has been recently proposed that can learn a third action-value head on a fixed dataset the same as for two-head net.
Hill Climbing on Value Estimates for Search-control in Dyna
In this work, we propose to generate such states by using the trajectory obtained from Hill Climbing (HC) the current estimate of the value function.
Distributional Reinforcement Learning for Efficient Exploration
In distributional reinforcement learning (RL), the estimated distribution of value function models both the parametric and intrinsic uncertainties.
Single-step Options for Adversary Driving
In this paper, we use reinforcement learning for safety driving in adversary settings.
Deep Reinforcement Learning with Decorrelation
Further experiments on the losing games show that our decorelation algorithms can win over DQN and QR-DQN with a fined tuned regularization factor.
ACE: An Actor Ensemble Algorithm for Continuous Control with Tree Search
In this paper, we propose an actor ensemble algorithm, named ACE, for continuous control with a deterministic policy in reinforcement learning.
QUOTA: The Quantile Option Architecture for Reinforcement Learning
In this paper, we propose the Quantile Option Architecture (QUOTA) for exploration based on recent advances in distributional reinforcement learning (RL).
Negative Log Likelihood Ratio Loss for Deep Neural Network Classification
In deep neural network, the cross-entropy loss function is commonly used for classification.
Practical Issues of Action-conditioned Next Image Prediction
Our major finding is that action tiling encoding is the most important factor leading to the remarkable performance of the CDNA model.
Universal Option Models
We prove that the UOM of an option can construct a traditional option model given a reward function, and the option-conditional return is computed directly by a single dot-product of the UOM with the reward function.
Multi-Step Dyna Planning for Policy Evaluation and Control
We extend Dyna planning architecture for policy evaluation and control in two significant aspects.
