Methods › General › Initialization › Fixup Initialization
Fixup Initialization
Introduced by Hongyi Zhang et al. in Fixup Initialization: Residual Learning Without Normalization
archive 2025-07-28 Description, source and code snippet are the archive's method entry.
FixUp Initialization, or Fixed-Update Initialization, is an initialization method that rescales the standard initialization of residual branches by adjusting for the network architecture. Fixup aims to enables training very deep residual networks stably at a maximal learning rate without normalization.
The steps are as follows:
-
Initialize the classification layer and the last layer of each residual branch to 0.
-
Initialize every other layer using a standard method, e.g. Kaiming Initialization, and scale only the weight layers inside residual branches by L^(1/(2m-2)).
-
Add a scalar multiplier (initialized at 1) in every branch and a scalar bias (initialized at 0) before each convolution, linear, and element-wise activation layer.
In Syntology name not resolved (unknown_name). Other entries in Syntology's graph carry this name; the lookup's note below says what that means for this name.
From the lookup's note on this name: Nothing resolved to this name, but 2 entries in Syntology's graph carry it; the most cited is the paper “Fixup Initialization: Residual Learning Without Normalization” (arXiv:1901.09321, 416 citations). The lookup reads this as a miss of its own name matching, not a gap in coverage: the graph holds the name under a spelling the lookup did not reach. For arXiv:1901.09321: its page on this site · ask the lookup about it.
Near matches the lookup lists, each a different name: Query Initialization; initialization memory; MAML co-initialization; centroid initialization; Lyapunov initialization; initialization algorithm; Grid-Based Initialization; Flow-Induced Initialization.
Syntology's have check for this name (free, it answers coverage only), read 2026-10-05; ask it again live. A level is the served verification level of that one answer; nothing here is summed across methods.
Papers archive 2025-07-28
archive papers tagged: 2 · with a code link: 1 · where Syntology ran a sample: 1 (1 with a run with no instrument failure, 0 where every run was a failure of Syntology's instrument) Syntology
Show: all tagged papersonly where code ran (1 of 2 tagged: 1 with a run with no instrument failure, 0 where every run was a failure of Syntology's instrument)
Page 1 of 1: papers 1 to 2 of 2, newest first by the archive's date (ties by slug), in archive order.
Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code, as “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”; the instrument figure counts failures of Syntology's instrument, not of the code. It is per sample and not a correctness claim. When the archive marks a repository official for the paper, the line starts with that repository's state (the archive's flag, not a verdict on who wrote the code; “community repositories only” when every sample that ran came from a community repository, “official: no sample here; runs from other or unrecorded repositories” when some came from a repository the paper names or has in its text, or from none recorded); hover it for the repositories the samples that ran came from.
-
Towards Efficient Full 8-bit Integer DNN Online Training on Resource-limited Devices without Batch Normalization 27 May 2021 · 0 repositories · arXiv:2105.13890
-
Fixup Initialization: Residual Learning Without Normalization 27 Jan 2019 · 10 repositories · arXiv:1901.09321Syntology 2 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; 1 where Syntology's instrument failed) · 2 unverified (of 4 harvested samples) · 4 pointer-only (licence)
Tasks archive 2025-07-28
7 tasks the archive attaches to papers tagged with this method, by distinct papers. A task without a page in the catalog is plain text.
| Task | Papers |
|---|---|
| General Classification | 1 |
| Image Classification | 1 |
| Machine Translation | 1 |
| Model Compression | 1 |
| Quantization | 1 |
| Translation | 1 |
| image-classification | 1 |
Usage over time archive 2025-07-28
Components: the archive holds no method-to-method composition, so PwC's Components table cannot be rebuilt; the Papers list carries no Results column for the same reason (the archive does not join its leaderboard rows to method tags).
Categories archive 2025-07-28
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