Temporal Aware Deep Reinforcement Learning

09/05/2021
by   Deepak-George Thomas, et al.
0

The function approximators employed by traditional image based Deep Reinforcement Learning (DRL) algorithms usually lack a temporal learning component and instead focus on learning the spatial component. We propose a technique wherein both temporal as well as spatial components are jointly learned. Our tested was tested with a generic DQN and it outperformed it in terms of maximum rewards as well as sample complexity. This algorithm has implications in the robotics as well as sequential decision making domains.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
01/21/2022

Deep reinforcement learning under signal temporal logic constraints using Lagrangian relaxation

Deep reinforcement learning (DRL) has attracted much attention as an app...
research
07/25/2023

Deep Reinforcement Learning for Robust Goal-Based Wealth Management

Goal-based investing is an approach to wealth management that prioritize...
research
12/02/2020

Are Gradient-based Saliency Maps Useful in Deep Reinforcement Learning?

Deep Reinforcement Learning (DRL) connects the classic Reinforcement Lea...
research
04/14/2023

Robust Decision-Making in Spatial Learning: A Comparative Study of Successor Features and Predecessor Features Algorithms

Predictive map theory, one of the theories explaining spatial learning i...
research
11/13/2019

Buffer-aware Wireless Scheduling based on Deep Reinforcement Learning

In this paper, the downlink packet scheduling problem for cellular netwo...
research
04/02/2019

Learning a Partitioning Advisor with Deep Reinforcement Learning

Commercial data analytics products such as Microsoft Azure SQL Data Ware...
research
08/23/2020

DSP: A Differential Spatial Prediction Scheme for Comprehensive real industrial datasets

Inverse Distance Weighted models (IDW) have been widely used for predict...

Please sign up or login with your details

Forgot password? Click here to reset