SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1810.06801 · 2018

Quasi-hyperbolic momentum and Adam for deep learning

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 7 functions out of this paper's own repositories and ran 6 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
facebookresearch/qhoptim canonical 6 of 7
FunctionStatusWhere it lives
allclose Ran facebookresearch/qhoptim/test_qhoptim/tf/util.py
pointer only (licence: MIT) · get_code("5597447a89943bbe")
build_net Ran facebookresearch/qhoptim/test_qhoptim/tf/util.py
pointer only (licence: MIT) · get_code("8c61c8f36e03fd81")
call_if_callable Ran facebookresearch/qhoptim/qhoptim/tf/util.py
pointer only (licence: MIT) · get_code("f70286c6efb24e49")
from_pid Ran facebookresearch/qhoptim/qhoptim/common/param_conv.py
pointer only (licence: MIT) · get_code("85d5bd9c96b637ab")
from_robust_momentum Ran facebookresearch/qhoptim/qhoptim/common/param_conv.py
pointer only (licence: MIT) · get_code("f7a58c6038a0e8fa")
from_synthesized_nesterov Ran facebookresearch/qhoptim/qhoptim/common/param_conv.py
pointer only (licence: MIT) · get_code("e8d42db83273329b")
QHAdamW Not yet run facebookresearch/qhoptim/qhoptim/pyt/qhadam.py
pointer only (licence: MIT) · get_code("e6d8e5bc3f47df76")

Repositories linked to this paper

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

Abstract

Momentum-based acceleration of stochastic gradient descent (SGD) is widely used in deep learning. We propose the quasi-hyperbolic momentum algorithm (QHM) as an extremely simple alteration of momentum SGD, averaging a plain SGD step with a momentum step. We describe numerous connections to and identities with other algorithms, and we characterize the set of two-state optimization algorithms that QHM can recover. Finally, we propose a QH variant of Adam called QHAdam, and we empirically demonstrate that our algorithms lead to significantly improved training in a variety of settings, including a new state-of-the-art result on WMT16 EN-DE. We hope that these empirical results, combined with the conceptual and practical simplicity of QHM and QHAdam, will spur interest from both practitioners and researchers. Code is immediately available.

For agents

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

get_harvested_code_for_paper("1810.06801")
get_code_for_paper("1810.06801")
have("1810.06801")

Connect an agent — have() is free.