{"id":"W3090302053","doi":"10.1109/tcsi.2021.3066967","title":"CARLA: A Convolution Accelerator With a Reconfigurable and Low-Energy Architecture","year":2021,"lang":"en","type":"preprint","venue":"IEEE Transactions on Circuits and Systems I Regular Papers","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Convolutional neural network; Dram; Convolution (computer science); Architecture; Latency (audio); Computation; Efficient energy use; Computer architecture; Computer engineering; Residual neural network; Embedded system; Parallel computing; Computer hardware; Artificial intelligence; Algorithm; Artificial neural network; Telecommunications","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":{"n_in":0,"stratum":"aff_core","weight":5595.2375,"opus":{"tier":"OUT","genre":"empirical","about_ca":false,"confidence":"high","reason":"Hardware architecture for convolutional neural network acceleration."},"gpt":{"tier":"OUT","genre":"empirical","about_ca":false,"confidence":"high","reason":"The paper develops a neural-network hardware accelerator rather than studying research practice."},"grok":{"tier":"OUT","genre":"empirical","about_ca":false,"confidence":"high","reason":"Hardware architecture for CNN acceleration, computer engineering domain work."}},"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.000213712,0.00062742,0.0003273471,0.0004893027,0.0003307958,0.0006235773,0.001567875,0.0004909774,0.006463596],"category_scores_gemma":[0.0004989923,0.0003128549,0.0004502319,0.0005167422,0.0003005318,0.0008159904,0.0007079624,0.0008420793,0.001856208],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008071589,"about_ca_system_score_gemma":0.001731233,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004789782,"about_ca_topic_score_gemma":0.008356786,"domain_scores_codex":[0.9997901,0.00001616069,0.00001146568,0.00004328102,0.00008485026,0.00005413399],"domain_scores_gemma":[0.9998325,0.00002965171,0.00002218164,0.00003567331,0.00005279535,0.00002714494],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.002737113,0.0005861951,0.006327671,0.0008188209,0.0003832839,0.0008992294,0.0001958744,0.1349926,0.2513276,0.03418207,0.08989326,0.4776562],"study_design_scores_gemma":[0.0003032966,0.0006509679,0.001783782,0.00005579901,0.0001169544,0.0006474185,0.00005402411,0.8088107,0.1244802,0.005544457,0.05747087,0.00008154075],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.2272686,0.003154726,0.6718002,0.001171054,0.001029006,0.0003734446,0.001487082,0.03986192,0.05385402],"genre_scores_gemma":[0.7095516,0.0007254534,0.2501727,0.0006592672,0.0000906397,0.000257451,0.002606259,0.0007427188,0.03519396],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.006463596,"threshold_uncertainty_score":0,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01462263658816678,"score_gpt":0.2101305978701999,"score_spread":0.1955079612820331,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}