Papers › Auto-NBA: Efficient and Effective Search Over the Joint Space of Networks, Bitwidths,...

Auto-NBA: Efficient and Effective Search Over the Joint Space of Networks, Bitwidths, and Accelerators

11 Jun 2021arXiv:2106.06575archive 2025-07-28

Yonggan Fu, Yongan Zhang, Yang Zhang, David Cox, Yingyan Celine Lin

While maximizing deep neural networks' (DNNs') acceleration efficiency requires a joint search/design of three different yet highly coupled aspects, including the networks, bitwidths, and accelerators, the challenges associated with such a joint search have not yet been fully understood and addressed. The key challenges include (1) the dilemma of whether to explode the memory consumption due to the huge joint space or achieve sub-optimal designs, (2) the discrete nature of the accelerator design space that is coupled yet different from that of the networks and bitwidths, and (3) the chicken and egg problem associated with network-accelerator co-search, i.e., co-search requires operation-wise hardware cost, which is lacking during search as the optimal accelerator depending on the whole network is still unknown during search. To tackle these daunting challenges towards optimal and fast development of DNN accelerators, we propose a framework dubbed Auto-NBA to enable jointly searching for the Networks, Bitwidths, and Accelerators, by efficiently localizing the optimal design within the huge joint design space for each target dataset and acceleration specification. Our Auto-NBA integrates a heterogeneous sampling strategy to achieve unbiased search with constant memory consumption, and a novel joint-search pipeline equipped with a generic differentiable accelerator search engine. Extensive experiments and ablation studies validate that both Auto-NBA generated networks and accelerators consistently outperform state-of-the-art designs (including co-search/exploration techniques, hardware-aware NAS methods, and DNN accelerators), in terms of search time, task accuracy, and accelerator efficiency. Our codes are available at: https://github.com/RICE-EIC/Auto-NBA.

PaperPDFCodeCode Syntology ranMethod as code

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 26 code samples: 24 from 1 repository linked to this paper, 2 identical to code harvested from another paper's repository. Across all 26: 18 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 17 where Syntology's instrument failed) · 8 unverified. Where Syntology's instrument failed: 4 where our contract draft was wrong; 13 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 24 samples from 1 repository, 16 ran (1 with no instrument failure; 15 where Syntology's instrument failed); 2 identical to code first harvested elsewhere. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

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

RICE-EIC/Auto-NBA officialmentioned in papermentioned on GitHubpytorchSyntology: 24 samples, 16 ran (15 instrument-failed) · 8 unverified 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.

Code Syntology ran Syntology

26 samples harvested: 18 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 17 where Syntology's instrument failed) · 8 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.

1ran · honoured contract
4ran · our draft was wrong
13ran; the contract check could not run
8unverified

Licence: 2 of the 26 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 RICE-EIC/Auto-NBA. 2 samples are listed last, in a group without a repository. “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. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

RICE-EIC/Auto-NBA official repository 24 samples, 16 ran (15 instrument-failed) · 8 unverified
allocate_layers model_search.py ran; the contract check could not run (RAISES) MIT (permissive) · dde6796a9649b047 · report fixture: signature net_struct:list/1/list, quant_info:list/1/int, dw:list/1/bool, platform_specs:dict/0/?, cifar:bool/0/bool · returned tuple · level recorded 2026-09-22
combined_latency model_search.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · 2745222d0ad20aeb · report fixture: signature comp_mode:int/0/int, input_params:list/1/float, net_struct:list/1/float, quant:int/0/int · returned float · level recorded 2026-09-22
comp_engine_lat model_search.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · d6f30bbd3dc5ba42 · report fixture: signature comp_mode:int/0/int, input_params:list/1/float, net_struct:list/1/float · returned float · level recorded 2026-09-22
design_choice_gen model_search.py ran · our draft was wrong fingerprinted MIT (permissive) · d4db52f918512b42 · report fixture: signature cifar:bool/0/bool, edd:bool/0/bool · returned tuple · level recorded 2026-09-22
dw_combined_latency model_search.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · 5d5557669b20df95 · report fixture: signature comp_mode:int/0/int, input_params:list/1/int, net_struct:list/1/int, quant:int/0/int · returned float · level recorded 2026-09-22
dw_comp_engine_lat model_search.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · d4f83a7f576254e7 · report fixture: signature comp_mode:int/0/int, input_params:list/1/int, net_struct:list/1/int · returned float · level recorded 2026-09-22
dw_read_if_lat model_search.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · 69a5f7e7f77884eb · report fixture: signature comp_mode:int/0/int, input_params:list/1/int, net_struct:list/1/int, quant:int/0/int · returned float · level recorded 2026-09-22
dw_read_we_lat model_search.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · 11ad6ba24dd52bed · report fixture: signature comp_mode:int/0/int, input_params:list/1/int, net_struct:list/1/int, quant:int/0/int · returned float · level recorded 2026-09-22
dw_write_ofmap model_search.py ran; the contract check could not run (RAISES) MIT (permissive) · 6915888ee6046399 · report fixture: signature comp_mode:int/0/int, input_params:list/1/int, net_struct:dict/0/?, quant:int/0/int · returned float · level recorded 2026-09-22
generate_params model_search.py ran · our draft was wrong MIT (permissive) · 1bab1f2c8a46c6fe · report fixture: signature acc1_space:dict/0/?, acc2_space:dict/0/?, dw_acc1_space:dict/0/?, dw_acc2_space:dict/0/?, quant_options:list/1/str · returned tuple · level recorded 2026-09-22
gumbel_softmax in 82 papers
census: 19 implementations from 20 papers, 12 distinct outputs across 2 buckets, one shared input per bucket this sample: bucket 1 of 2, arg 1: rank 2, kind float, dtype float32
model_search.py ran · our draft was wrong MIT (permissive) · 1824fbfcfee2e185 · report fixture: returned Tensor · level recorded 2026-09-22
random_sample in 18 papers model_search.py ran · honoured contract MIT (permissive) · 40eb5ef310a13593 · report fixture: signature input_dict:dict/0/?, fix_comp_mode:bool/0/bool · returned list · level recorded 2026-09-22
read_if_lat model_search.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · 2cf026e1cb51cdf0 · report fixture: signature comp_mode:int/0/int, input_params:list/1/int, net_struct:list/1/int, quant:int/0/int · returned float · level recorded 2026-09-22
read_we_lat model_search.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · b678c5ea7942bb72 · report fixture: signature comp_mode:int/0/int, input_params:list/1/int, net_struct:list/1/int, quant:int/0/int · returned float · level recorded 2026-09-22
resource_consumption model_search.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · 377dd9a088043ca5 · report fixture: signature comp_mode:int/0/int, input_params:list/1/int, net_struct:list/1/int, dw:bool/0/bool, quant:int/0/int · returned tuple · level recorded 2026-09-22
write_ofmap model_search.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · 082bac1336b391d0 · report fixture: signature comp_mode:int/0/int, input_params:list/1/int, net_struct:NoneType/0/?, quant:int/0/int · returned float · level recorded 2026-09-22
FBNet model_search.py unverified MIT (permissive) · 7809544871a9fe0b · report
MixedOp in 8 papers model_search.py unverified MIT (permissive) · 7e63cdc000697a3c · report
capsuled_predictor model_search.py unverified MIT (permissive) · 83866a815206c036 · report
cifar_convert_to_layers model_search.py unverified MIT (permissive) · 9e23744959c78f65 · report
evaluate_latency model_search.py unverified MIT (permissive) · bdf35073bdcf40de · report
search_for_best_latency model_search.py unverified MIT (permissive) · bd71a08555ee6e6c · report
sys_consumption model_search.py unverified MIT (permissive) · cf3df44ac78c9623 · report
sys_latency model_search.py unverified MIT (permissive) · f871981fcbaf0534 · report
Repository not located 2 samples, 2 ran (2 instrument-failed) · 0 unverified
2 samples are identical to code Syntology harvested from another paper's repository; this paper's own copy was not located, so no repository is named and no licence is recorded.
gumbel_softmax_sample in 49 papers
census: 14 implementations from 16 papers, 4 distinct outputs on one shared input this sample: bucket 1 of 1, arg 1: rank 2, kind float, dtype float32
repository not located ran; the contract check could not run (RAISES) licence of this copy not recorded · 0d41bbe52278fde1 · report fixture: returned Tensor · level recorded 2026-09-22
sample_gumbel in 62 papers repository not located ran · our draft was wrong licence of this copy not recorded · 7c913c414af7196f · report fixture: returned Tensor · level recorded 2026-09-22

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it.

1 assembly on this paper: 1 harvested, 0 generated.

harvested Assembly for Auto-NBA (graph method: Auto-NBA). 26 bricks; individually on a fixture: 18 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 17 where Syntology's instrument failed) · 8 unverified. The assembly has not been run end-to-end.

  1. FBNet entry · level 0 · unverified · 7809544871a9fe0b
  2. MixedOp model · level 0 · unverified · 7e63cdc000697a3c
  3. evaluate_latency helper · level 0 · unverified · bdf35073bdcf40de
  4. gumbel_softmax sampler · level 1 · ran · our draft was wrong · 1824fbfcfee2e185
  5. search_for_best_latency scheduler · level 0 · unverified · bd71a08555ee6e6c
  6. capsuled_predictor helper · level 0 · unverified · 83866a815206c036
  7. gumbel_softmax_sample helper · level 1 · ran; the contract check could not run (RAISES) · 0d41bbe52278fde1
  8. design_choice_gen helper · level 1 · ran · our draft was wrong · d4db52f918512b42
  9. generate_params helper · level 1 · ran · our draft was wrong · 1bab1f2c8a46c6fe
  10. random_sample helper · level 1 · ran · honoured contract · 40eb5ef310a13593
  11. allocate_layers helper · level 1 · ran; the contract check could not run (RAISES) · dde6796a9649b047
  12. cifar_convert_to_layers helper · level 0 · unverified · 9e23744959c78f65
  13. sys_consumption helper · level 0 · unverified · cf3df44ac78c9623
  14. sys_latency helper · level 0 · unverified · f871981fcbaf0534
  15. sample_gumbel sampler · level 1 · ran · our draft was wrong · 7c913c414af7196f
  16. resource_consumption helper · level 1 · ran; the contract check could not run (RAISES) · 377dd9a088043ca5
  17. combined_latency helper · level 1 · ran; the contract check could not run (RAISES) · 2745222d0ad20aeb
  18. dw_combined_latency helper · level 1 · ran; the contract check could not run (RAISES) · 5d5557669b20df95
  19. comp_engine_lat helper · level 1 · ran; the contract check could not run (RAISES) · d6f30bbd3dc5ba42
  20. read_if_lat helper · level 1 · ran; the contract check could not run (RAISES) · 2cf026e1cb51cdf0
  21. read_we_lat helper · level 1 · ran; the contract check could not run (RAISES) · b678c5ea7942bb72
  22. write_ofmap helper · level 1 · ran; the contract check could not run (RAISES) · 082bac1336b391d0
  23. dw_comp_engine_lat helper · level 1 · ran; the contract check could not run (RAISES) · d4f83a7f576254e7
  24. dw_read_if_lat helper · level 1 · ran; the contract check could not run (RAISES) · 69a5f7e7f77884eb
  25. dw_read_we_lat helper · level 1 · ran; the contract check could not run (RAISES) · 11ad6ba24dd52bed
  26. dw_write_ofmap helper · level 1 · ran; the contract check could not run (RAISES) · 6915888ee6046399

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2106.06575") returns these assemblies with each brick's role and wiring (how to connect).

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

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