SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2303.17589 · ICML · 2023

Polarity is all you need to learn and transfer faster

Joshua Vogelstein, Qingyang Wang, Michael Powell, Eric Bridgeford, Ali Geisa

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 13 functions out of this paper's own repositories and ran 12 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
aliceqingyangwang/weightpolarityexpr canonical 12 of 13
FunctionStatusWhere it lives
build_string_from_dict Ran aliceqingyangwang/weightpolarityexpr/CV/helperfun.py
code served (permissive licence) · get_code("8850801905ebce00")
getSingleLayerNet Ran aliceqingyangwang/weightpolarityexpr/XOR/getSingleLayerNet.py
code served (permissive licence) · get_code("e9dafb67fa011eea")
get_config Ran aliceqingyangwang/weightpolarityexpr/CV/batch_by_run.py
code served (permissive licence) · get_code("1ed2f018d790cd11")
get_config Ran aliceqingyangwang/weightpolarityexpr/XOR/batch_by_run.py
code served (permissive licence) · get_code("23ec93e259d03ad1")
get_dict_by_key Ran aliceqingyangwang/weightpolarityexpr/CV/helperfun.py
code served (permissive licence) · get_code("a71e24e23ab4b237")
get_train_param Ran aliceqingyangwang/weightpolarityexpr/CV/tf_training.py
code served (permissive licence) · get_code("ba8083f19762e1ab")
plot_diff_plus_mannwhitneyu Ran aliceqingyangwang/weightpolarityexpr/CV/plots.py
code served (permissive licence) · get_code("90b725ddfaeb0813")
plot_median_plus_example Ran aliceqingyangwang/weightpolarityexpr/CV/plots.py
code served (permissive licence) · get_code("f2aa94917445b090")
plot_sem Ran aliceqingyangwang/weightpolarityexpr/CV/plots.py
code served (permissive licence) · get_code("9b95d7f67d7b9d13")
plot_weights Ran aliceqingyangwang/weightpolarityexpr/CV/helperfun.py
code served (permissive licence) · get_code("f8b5c23e253c996e")
train_model Ran aliceqingyangwang/weightpolarityexpr/CV/tf_training.py
code served (permissive licence) · get_code("d0012529718bb0d0")
update_job_list Ran aliceqingyangwang/weightpolarityexpr/CV/scheduler.py
code served (permissive licence) · get_code("3ffd16006acd8441")
define_model Not yet run aliceqingyangwang/weightpolarityexpr/CV/tf_training.py
code served (permissive licence) · get_code("e5c502cca4dcb7b4")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Natural intelligences (NIs) thrive in a dynamic world -they learn quickly, sometimes with only a few samples. In contrast, artificial intelligences (AIs) typically learn with a prohibitive number of training samples and computational power. What design principle difference between NI and AI could contribute to such a discrepancy? Here, we investigate the role of weight polarity: development processes initialize NIs with advantageous polarity configurations; as NIs grow and learn, synapse magnitudes update, yet polarities are largely kept unchanged. We demonstrate with simulation and image classification tasks that if weight polarities are adequately set a priori, then networks learn with less time and data. We also explicitly illustrate situations in which a priori setting the weight polarities is disadvantageous for networks. Our work illustrates the value of weight polarities from the perspective of statistical and computational efficiency during learning.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2303.17589")
get_code_for_paper("2303.17589")
have("2303.17589")

Connect an agent — have() is free.