Atari River Raid Environment

Overview

Viewed from a top-down perspective, the player flies a fighter jet over the River of No Return in a raid behind enemy lines. The player’s jet can only move left and right—it cannot maneuver up and down the screen—but it can accelerate and decelerate. The player’s jet crashes if it collides with the riverbank or an enemy craft, or if the jet runs out of fuel. Assuming fuel can be replenished, and if the player evades damage, gameplay is essentially unlimited.

The player scores points for shooting enemy tankers (30 pts), helicopters (60 pts), fuel depots (80 pts), jets (100 pts), and bridges (500 pts). The jet refuels when it flies over a fuel depot. A bridge marks the end of a game level. Non-Atari 2600 ports of the game add hot air balloons that are worth 60 points when shot as well as tanks along the sides of the river that shoot at the player’s jet.

Destroying bridges also serve as the game’s checkpoints. If the player crashes the plane they will start their next life at the last destroyed bridge.

Description from Wikipedia

Performances of RL Agents

We list various reinforcement learning algorithms that were tested in this environment. These results are from RL Database. If this page was helpful, please consider giving a star!

Star

Human Starts

Result Algorithm Source
18184.4 Prioritized DDQN (prop, tuned) Prioritized Experience Replay
16569.4 DuDQN Dueling Network Architectures for Deep Reinforcement Learning
16496.8 PDD DQN Dueling Network Architectures for Deep Reinforcement Learning
14382.2 Human Massively Parallel Methods for Deep Reinforcement Learning
12201.8 A3C FF Asynchronous Methods for Deep Reinforcement Learning
11807.2 Prioritized DDQN (rank, tuned) Prioritized Experience Replay
10838.4 DDQN (tuned) Deep Reinforcement Learning with Double Q-learning
10205.5 Prioritized DQN (rank) Prioritized Experience Replay
10001.2 A3C FF 1 day Asynchronous Methods for Deep Reinforcement Learning
6591.9 A3C LSTM Asynchronous Methods for Deep Reinforcement Learning
6579.0 DDQN Deep Reinforcement Learning with Double Q-learning
5310.27 Gorila DQN Massively Parallel Methods for Deep Reinforcement Learning
4065.3 DQN Massively Parallel Methods for Deep Reinforcement Learning
588.3 Random Massively Parallel Methods for Deep Reinforcement Learning

No-op Starts

Result Algorithm Source
29608.05 IMPALA (deep) IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures
23134 NoisyNet DuDQN Noisy Networks for Exploration
21162.6 DDQN A Distributional Perspective on Reinforcement Learning
21162.6 DuDQN Dueling Network Architectures for Deep Reinforcement Learning
20607.6 PDD DQN Dueling Network Architectures for Deep Reinforcement Learning
18405 DuDQN Noisy Networks for Exploration
17765 IQN Implicit Quantile Networks for Distributional Reinforcement Learning
17762.8 ACKTR Scalable trust-region method for deep reinforcement learning using Kronecker-factored approximation
17571 QR-DQN-1 Distributional Reinforcement Learning with Quantile Regression
17401.9 IMPALA (shallow) IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures
17380.7 Reactor The Reactor: A fast and sample-efficient Actor-Critic agent for Reinforcement Learning
17322 C51 A Distributional Perspective on Reinforcement Learning
17118.0 Human Dueling Network Architectures for Deep Reinforcement Learning
16957.3 Reactor ND The Reactor: A fast and sample-efficient Actor-Critic agent for Reinforcement Learning
16608.3 Reactor The Reactor: A fast and sample-efficient Actor-Critic agent for Reinforcement Learning
13513.3 Human Human-level control through deep reinforcement learning
12015.3 DDQN Deep Reinforcement Learning with Double Q-learning
9425 NoisyNet DQN Noisy Networks for Exploration
9336 QR-DQN-0 Distributional Reinforcement Learning with Quantile Regression
8344.83 Gorila DQN Massively Parallel Methods for Deep Reinforcement Learning
8316 DQN Human-level control through deep reinforcement learning
8135 A3C Noisy Networks for Exploration
7878 NoisyNet A3C Noisy Networks for Exploration
7377.6 DQN A Distributional Perspective on Reinforcement Learning
7241 DQN Noisy Networks for Exploration
2850.15 IMPALA (deep, multitask) IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures
2650 Contingency Human-level control through deep reinforcement learning
1904 Linear Human-level control through deep reinforcement learning
1338.5 Random Human-level control through deep reinforcement learning

Normal Starts

Result Algorithm Source
9125.1 ACER Proximal Policy Optimization Algorithm
8393.6 PPO Proximal Policy Optimization Algorithm
7653.5 A2C Proximal Policy Optimization Algorithm