TF-Agents
E97077
TF-Agents is an open-source library built on TensorFlow that provides modular components and tools for developing, training, and evaluating reinforcement learning algorithms.
All labels observed (2)
| Label | Occurrences |
|---|---|
| TF-Agents canonical | 1 |
| TensorFlow Agents | 1 |
How this entity was disambiguated
This entity first appeared as the object of triple T816547 — resolving that mention is where its identity was fixed. The disambiguator weighed these candidate entities and picked the highlighted one (or “None”, minting a new entity). This is how homonymy is resolved: the same surface form can point to different entities.
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: TF-Agents Context triple: [TensorFlow, hasComponent, TF-Agents]
-
A.
OpenAI Baselines
OpenAI Baselines is a collection of high-quality reference implementations of reinforcement learning algorithms released by OpenAI for research and benchmarking.
-
B.
RLlib
RLlib is a scalable, open-source reinforcement learning library built on Ray that provides high-level APIs and distributed training support for a wide range of RL algorithms.
-
C.
OpenAI Gym
OpenAI Gym is a toolkit for developing and comparing reinforcement learning algorithms through a standardized collection of environments and interfaces.
-
D.
MuZero
MuZero is a DeepMind reinforcement learning algorithm that learns to plan and master complex games like Go, chess, and Atari without being given the rules in advance.
-
E.
DeepMind
DeepMind is a leading artificial intelligence research company renowned for breakthroughs such as AlphaGo and deep reinforcement learning, operating as a subsidiary of Google.
- F. None of above. chosen
- G. Unsure - the case is ambiguous/there is not enough information to decide.
NED2
Entity disambiguation (via description)
gpt-5-mini-2025-08-07
Target entity: TF-Agents Target entity description: TF-Agents is an open-source library built on TensorFlow that provides modular components and tools for developing, training, and evaluating reinforcement learning algorithms.
-
A.
OpenAI Baselines
OpenAI Baselines is a collection of high-quality reference implementations of reinforcement learning algorithms released by OpenAI for research and benchmarking.
-
B.
RLlib
RLlib is a scalable, open-source reinforcement learning library built on Ray that provides high-level APIs and distributed training support for a wide range of RL algorithms.
-
C.
OpenAI Gym
OpenAI Gym is a toolkit for developing and comparing reinforcement learning algorithms through a standardized collection of environments and interfaces.
-
D.
MuZero
MuZero is a DeepMind reinforcement learning algorithm that learns to plan and master complex games like Go, chess, and Atari without being given the rules in advance.
-
E.
DeepMind
DeepMind is a leading artificial intelligence research company renowned for breakthroughs such as AlphaGo and deep reinforcement learning, operating as a subsidiary of Google.
- F. None of above. chosen
Statements (49)
| Predicate | Object |
|---|---|
| instanceOf |
open-source project
ⓘ
reinforcement learning framework ⓘ software library ⓘ |
| basedOn | TensorFlow ⓘ |
| designedFor |
production reinforcement learning systems
ⓘ
research in reinforcement learning ⓘ |
| developedBy |
Google
ⓘ
Google Brain ⓘ
surface form:
Google Brain team
|
| hasComponent |
agents
ⓘ
bandits library ⓘ drivers ⓘ environments ⓘ metrics ⓘ networks ⓘ policies ⓘ replay buffers ⓘ |
| hasFeature |
data collection drivers
ⓘ
distributional RL support ⓘ experience replay ⓘ multi-armed bandits support ⓘ |
| hostedOn | GitHub ⓘ |
| isOpenSource | true ⓘ |
| license | Apache License 2.0 ⓘ |
| programmingLanguage | Python ⓘ |
| provides |
documentation
ⓘ
example notebooks ⓘ modular components ⓘ tools for developing reinforcement learning algorithms ⓘ tools for evaluating reinforcement learning algorithms ⓘ tools for training reinforcement learning algorithms ⓘ |
| supports |
TensorFlow
ⓘ
surface form:
TensorFlow 2
continuous action spaces ⓘ discrete action spaces ⓘ eager execution ⓘ off-policy algorithms ⓘ on-policy algorithms ⓘ reinforcement learning ⓘ tf.function graphs ⓘ |
| supportsAlgorithmFamily |
DDPG
ⓘ
Atari deep Q-network ⓘ
surface form:
DQN
PPO ⓘ REINFORCE ⓘ SAC ⓘ TD3 ⓘ actor-critic methods ⓘ policy gradient methods ⓘ |
| supportsEnvironment |
Atari environments
ⓘ
MuJoCo environments ⓘ OpenAI Gym ⓘ |
How these facts were elicited
The pipeline generated the facts above by prompting gpt-5.1 with this entity's name + description and the instruction below.
Instruction
You are a knowledge base construction expert. Given a subject entity and a description of it, return factual statements that you know for the subject as a JSON list of dictionaries(triples), where keys must be "subject", "predicate" and "object". The number of facts may be very high, between 25 to 50 or more, for very popular subjects. For less popular subjects, the number of facts can be very low, like 5 or 10. # Requirements - If you don't know the subject at all, return an empty list. - If the subject is not a named entity, return an empty list. - Include at least one triple where predicate is "instanceOf". - Do not get too wordy. - Separate several objects into multiple triples with one object.
Input
Subject: TF-Agents Description of subject: TF-Agents is an open-source library built on TensorFlow that provides modular components and tools for developing, training, and evaluating reinforcement learning algorithms.
Referenced by (2)
Full triples — surface form annotated when it differs from this entity's canonical label.
this entity surface form:
TensorFlow Agents