Pith Number

pith:V4AFOVJK

pith:2015:V4AFOVJKLCVFYXBPU3X53RSMDZ

not attested not anchored not stored refs resolved

Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift

Christian Szegedy, Sergey Ioffe

Batch Normalization normalizes each layer's inputs using mini-batch statistics, allowing higher learning rates and faster convergence in deep networks.

arxiv:1502.03167 v3 · 2015-02-11 · cs.LG

Open paper page JSON Open Graph Bundle Merged state Verified badge What is a Pith Number?

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{V4AFOVJKLCVFYXBPU3X53RSMDZ}

Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge

Record completeness

1 Bitcoin timestamp

2 Internet Archive

3 Author claim open · sign in to claim

4 Citations open

5 Replications open

✓ Portable graph bundle live · download bundle · merged state

The bundle contains the canonical record plus signed events. A mirror can host it anywhere and recompute the same current state with the deterministic merge algorithm.

Claims

C1strongest claim

Batch Normalization allows us to use much higher learning rates and be less careful about initialization. It also acts as a regularizer, in some cases eliminating the need for Dropout. Applied to a state-of-the-art image classification model, Batch Normalization achieves the same accuracy with 14 times fewer training steps, and beats the original model by a significant margin.

C2weakest assumption

That the changing distribution of each layer's inputs (internal covariate shift) is the main cause of slow training and that normalizing per mini-batch will reliably reduce this shift without introducing new instabilities or requiring extensive additional tuning.

C3one line summary

Batch Normalization normalizes layer inputs per mini-batch to reduce internal covariate shift, allowing higher learning rates, less careful initialization, and faster convergence in deep networks.

References

24 extracted · 24 resolved · 1 Pith anchors

[1] Understanding the difficulty of training deep feedforward neural networks 2010

[2] Large scale distributed deep networks 2012

[3] Natural neural networks

[4] Adaptive subgradient methods for online learning and stochastic optimization 2011

[5] Knowledge matters: Importance of prior information for optimization 2013 · arXiv:1301.4083

Formal links

2 machine-checked theorem links

Cited by

94 papers in Pith

SingGuard: A Policy-Adaptive Multimodal LLM Guardrail with Dynamic Reasoning

Acceleration of an algebraic multigrid pressure solver using graph neural networks

Higher-order effects in amplitude-assisted polarisation extraction with machine-learning techniques

SingGuard: A Policy-Adaptive Multimodal LLM Guardrail with Dynamic Reasoning

Explaining Machine Learning and Memorization with Statistical Mechanics

Receipt and verification

First computed	2026-07-04T19:21:06.750810Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

af0057552a58aa5c5c2fa6efddc64c1e5f8fb19d65deeae32e474604e21b1978

Aliases

arxiv: 1502.03167 · arxiv_version: 1502.03167v3 · doi: 10.48550/arxiv.1502.03167 · pith_short_12: V4AFOVJKLCVF · pith_short_16: V4AFOVJKLCVFYXBP · pith_short_8: V4AFOVJK

Agent API

Resolver JSON Graph JSON Events JSON Schema Signing key

Verify this Pith Number yourself

curl -sH 'Accept: application/ld+json' https://pith.science/pith/V4AFOVJKLCVFYXBPU3X53RSMDZ \
  | jq -c '.canonical_record' \
  | python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: af0057552a58aa5c5c2fa6efddc64c1e5f8fb19d65deeae32e474604e21b1978

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "60a191248d53ddb724f599878ce1e47e7e12af75a8bf68a73d00fdf684b1c2d6",
    "cross_cats_sorted": [],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.LG",
    "submitted_at": "2015-02-11T01:44:18Z",
    "title_canon_sha256": "0f2bb60db4605b8b1026d18548cc20a23662a3a1772aaa7318839ddcd1ae4354"
  },
  "schema_version": "1.0",
  "source": {
    "id": "1502.03167",
    "kind": "arxiv",
    "version": 3
  }
}