Papers › Soft Actor-Critic for Discrete Action Settings

Soft Actor-Critic for Discrete Action Settings

16 Oct 2019arXiv:1910.07207archive 2025-07-28

Petros Christodoulou

Soft Actor-Critic is a state-of-the-art reinforcement learning algorithm for continuous action settings that is not applicable to discrete action settings. Many important settings involve discrete actions, however, and so here we derive an alternative version of the Soft Actor-Critic algorithm that is applicable to discrete action settings. We then show that, even without any hyperparameter tuning, it is competitive with the tuned model-free state-of-the-art on a selection of games from the Atari suite.

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 18 code samples harvested from 5 repositories linked to this paper: 12 ran (of which 0 constructed an object rather than computing a result; 12 with no instrument failure: 2 honoured, 1 violated, 9 with no contract checked; 0 where Syntology's instrument failed) · 6 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code; here there are none.

By repository: official repository: 2 samples from 1 repository, 0 ran (0 with no instrument failure; 0 where Syntology's instrument failed); community (archive-listed): 16 samples from 4 repositories, 12 ran (12 with no instrument failure; 0 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

No sample is linked to this paper from 2 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

13 repositories listed; official and paper-mentioned ones first.

p-christ/Deep-Reinforcement-Learning-Algorithms-with-PyTorch officialmentioned in papermentioned on GitHubpytorchMITSyntology: 2 samples, 0 ran · 2 unverified report
Bigpig4396/PyTorch-Soft-Actor-Critic-SAC mentioned on GitHubpytorchSyntology: not harvested report
Epsilon10/MARL mentioned on GitHubpytorchSyntology: not harvested report
ToolManChang/DRLseminar_code_challenge mentioned on GitHubSyntology: not harvested report
ac-93/soft-actor-critic mentioned on GitHubtfSyntology: 3 samples, 3 ran (0 instrument-failed) · 0 unverified report
cindycia/Atari-SAC-Discrete mentioned on GitHubpytorchMITSyntology: 7 samples, 5 ran (0 instrument-failed) · 2 unverified report
keraJLi/rejax mentioned on GitHubjaxApache-2.0Syntology: 5 samples, 4 ran (0 instrument-failed) · 1 unverified report
ku2482/rljax mentioned on GitHubjaxMITSyntology: not harvested report
ku2482/rltorch mentioned on GitHubpytorchMITSyntology: 1 sample, 0 ran · 1 unverified report
ku2482/sac-discrete.pytorch mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Prioritized Experience Replay (arXiv:1511.05952; that paper's own run record: 80 ran (of which 62 constructed an object rather than computing a result; 72 with no instrument failure: 4 honoured, 0 violated, 68 with no contract checked; 8 where Syntology's instrument failed) · 31 unverified); nothing checks that they implement this paper's method. report
lingweizhu/Pytorch-MunchausenActorCritic mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Munchausen Reinforcement Learning (arXiv:2007.14430; that paper's own run record: 12 ran (of which 11 constructed an object rather than computing a result; 12 with no instrument failure: 0 honoured, 1 violated, 11 with no contract checked; 0 where Syntology's instrument failed) · 10 unverified); nothing checks that they implement this paper's method. report
mindspore-courses/Deep-Reinforcement-Learning-Algorithms-with-MindSpore mentioned on GitHubmindsporeSyntology: not harvested report
BY571/SAC_discrete pytorchSyntology: not harvested report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

18 samples harvested: 12 ran (of which 0 constructed an object rather than computing a result; 12 with no instrument failure: 2 honoured, 1 violated, 9 with no contract checked; 0 where Syntology's instrument failed) · 6 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

2ran · honoured contract
1ran · violated contract
9ran
6unverified

Licence: 3 of the 18 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 5 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. 3 samples whose link to this paper records no repository are listed under one of this paper's own repositories, because identical code was harvested from it for another paper. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity.

p-christ/Deep-Reinforcement-Learning-Algorithms-with-PyTorch official repository 2 samples, 0 ran · 2 unverified
make_atari_game environments/Atari_Environment.py unverified MIT (permissive) · 5518bf3701f7ad8c · report
wrap_deepmind in 17 papers environments/Atari_Environment.py unverified MIT (permissive) · f22df3c0c0b99757 · report
ac-93/soft-actor-critic community (archive-listed) 3 samples, 3 ran (0 instrument-failed) · 0 unverified
process_action in 6 papers array_observation/sac_discrete/sac.py identical code Syntology harvested from this repository for another paper; this paper's link records no repository, path or licence of its own ran · honoured contract fingerprinted licence of this copy not recorded · e5cbade112f15d84 · report fixture: signature a:int/0/int, act_dim:int/0/int · returned ndarray · level recorded 2026-09-22
process_observation array_observation/sac_discrete/sac.py identical code Syntology harvested from this repository for another paper; this paper's link records no repository, path or licence of its own ran · honoured contract fingerprinted licence of this copy not recorded · 22e993f58f0d200b · report fixture: signature o:int/0/int, obs_dim:int/0/int, observation_type:str/0/str · returned ndarray · level recorded 2026-09-22
process_reward array_observation/sac_discrete/sac.py identical code Syntology harvested from this repository for another paper; this paper's link records no repository, path or licence of its own ran · violated contract fingerprinted licence of this copy not recorded · 1deaa4d231ac8f53 · report fixture: signature reward:Tensor/1/float · returned Tensor · level recorded 2026-09-22
cindycia/Atari-SAC-Discrete community (archive-listed) 7 samples, 5 ran (0 instrument-failed) · 2 unverified
create_dqn_base rltorch/network/builder.py ran MIT (permissive) · c010a6ac8668e20d · report fixture: signature num_channels:int/0/int, initializer:str/0/str · returned Sequential · level recorded 2026-09-24
create_linear_network rltorch/network/builder.py ran MIT (permissive) · 95498252faaec791 · report fixture: signature input_dim:int/0/int, output_dim:int/0/int, hidden_units:list/1/int, hidden_activation:str/0/str, output_activation:str/0/str, initializer:str/0/str · returned Sequential · level recorded 2026-09-24
initialize_weights in 30 papers rltorch/network/builder.py ran MIT (permissive) · a6efcc2f264c05c0 · report fixture: signature initializer:callable/0/? · returned function · level recorded 2026-09-24
make_atari in 17 papers rltorch/env/atari_wrappers.py ran MIT (permissive) · 4bc698f17fe89d85 · report fixture: signature env_id:str/0/str · returned MaxAndSkipEnv · level recorded 2026-09-24
to_batch rltorch/agent/utils.py ran MIT (permissive) · 3afd917c65b80e08 · report fixture: signature state:list/1/float, action:float/0/float, reward:float/0/float, next_state:list/1/float, done:float/0/float, device:device/0/? · returned tuple · level recorded 2026-09-24
make_pytorch_env rltorch/env/atari_wrappers.py unverified MIT (permissive) · 20aa2a394b00c4d7 · report
wrap_deepmind_pytorch rltorch/env/atari_wrappers.py unverified MIT (permissive) · 4d7ad6b7f3100923 · report
keraJLi/rejax community (archive-listed) 5 samples, 4 ran (0 instrument-failed) · 1 unverified
EpsilonGreedyPolicy src/rejax/networks.py ran Apache-2.0 (permissive) · 159ca677b48225e4 · report fixture: returned type · level recorded 2026-09-24
evaluate_single src/rejax/evaluate.py ran Apache-2.0 (permissive) · 098ce6dc4eff278c · report fixture: signature act:callable/0/?, env:MockEnv/0/?, env_params:MockEnvParams/0/?, rng:ArrayImpl/1/int, max_steps_in_episode:int/0/int · returned tuple · level recorded 2026-09-24
register_init src/rejax/algos/algorithm.py ran Apache-2.0 (permissive) · e62f6ecb2ba1c892 · report fixture: signature func:callable/0/? · returned function · level recorded 2026-09-24
update_rms src/rejax/algos/mixins.py ran Apache-2.0 (permissive) · ea614117ba22d252 · report fixture: signature rms_state:RMSState/0/?, x:ArrayImpl/2/float, batched:bool/0/bool · returned RMSState · level recorded 2026-09-24
EpsilonGreedyPolicy src/rejax/algos/iqn.py unverified Apache-2.0 (permissive) · 6fe3b3b62f1d627e · report
ku2482/rltorch community (archive-listed) 1 sample, 0 ran · 1 unverified
make_pytorch_env rltorch/env/atari_wrappers.py unverified MIT (permissive) · 4710cb8bbbee96dd · report

Tasks

Atari GamesReinforcement LearningReinforcement Learning (RL)reinforcement-learning

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Atari Games Atari 2600 Alien SAC Score 216.9 #47 of 49 Archive leaderboard report
Atari Games Atari 2600 Amidar SAC Score 7.9 #48 of 48 Archive leaderboard report
Atari Games Atari 2600 Assault SAC Score 350 #45 of 45 Archive leaderboard report
Atari Games Atari 2600 Asterix SAC Score 272 #49 of 49 Archive leaderboard report
Atari Games Atari 2600 Battle Zone SAC Score 4386.7 #46 of 47 Archive leaderboard report
Atari Games Atari 2600 Beam Rider SAC Score 432.1 #48 of 49 Archive leaderboard report
Atari Games Atari 2600 Breakout SAC Score 0.7 #58 of 58 Archive leaderboard report
Atari Games Atari 2600 Crazy Climber SAC Score 3668.7 #48 of 49 Archive leaderboard report
Atari Games Atari 2600 Enduro SAC Score 0.8 #44 of 48 Archive leaderboard report
Atari Games Atari 2600 Freeway SAC Score 4.4 #51 of 59 Archive leaderboard report
Atari Games Atari 2600 Frostbite SAC Score 59.4 #53 of 53 Archive leaderboard report
Atari Games Atari 2600 James Bond SAC Score 68.3 #44 of 45 Archive leaderboard report
Atari Games Atari 2600 Kangaroo SAC Score 29.3 #46 of 47 Archive leaderboard report
Atari Games Atari 2600 Ms. Pacman SAC Score 690.9 #45 of 47 Archive leaderboard report
Atari Games Atari 2600 Pong SAC Score -20.98 #52 of 52 Archive leaderboard report
Atari Games Atari 2600 Q*Bert SAC Score 280.5 #53 of 57 Archive leaderboard report
Atari Games Atari 2600 Road Runner SAC Score 305.3 #42 of 44 Archive leaderboard report
Atari Games Atari 2600 Seaquest SAC Score 211.6 #55 of 57 Archive leaderboard report
Atari Games Atari 2600 Space Invaders SAC Score 160.8 #54 of 55 Archive leaderboard report
Atari Games Atari 2600 Up and Down SAC Score 250.7 #44 of 44 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections