Tag
normalization
- Oct 3, 2026 Layer Normalization and the Residual Stream How per-example statistics change the gradient path compared with batch statistics, why the placement of a normalization layer relative to a residual branch alters scale growth with depth, and what the learnable gains do. deep learning
- Jul 1, 2026 Join Dependencies and the Ternary Structure Fifth Normal Form Protects Why a relation can be pairwise lossless and still not be a join of two projections, how a three-way join dependency arises from a cyclic constraint, and what 5NF removes. databases
- Jun 10, 2026 Boyce-Codd Normal Form and the Cost of Losing a Dependency Why BCNF decomposition can make a functional dependency unverifiable without a join, how 3NF avoids that, and when the extra anomaly is the better trade. databases
- Sep 18, 2025 Merging Profiles Across a Fleet Without Losing the Tail Why summing per-host profiles answers a different question than profiling the fleet, how weight normalization changes the answer, and what a merged profile cannot tell you. observability
- Aug 14, 2025 Multivalued Dependencies and the Redundancy BCNF Cannot See How an MVD asserts independence between two attribute groups, why the four-way combination of independent lists forces redundancy, and how 4NF decomposes on it. databases
- Jan 23, 2024 Attribute Closure and the Mechanical Derivation of Candidate Keys How Armstrong axioms generate the closure of a dependency set, why closure of an attribute set is computed by repeated scanning, and how to enumerate minimal keys without exponential search. databases
- Oct 15, 2023 Batch Normalization and the Train-Eval Gap Why batch statistics at training time and running averages at inference time produce different outputs, and how that mismatch is exploited to estimate what a network actually learned. deep learning
- Dec 30, 2022 Gradient Accumulation: What Micro-Batching Preserves and What It Does Not Splitting a batch into micro-batches keeps the arithmetic of a large batch but changes normalisation, batch-dependent layers, and the timing of every state update. deep learning
- Dec 7, 2021 Answer Normalization and the Fragility of Exact Match Exact-match scoring turns a string comparison into a quality verdict. The normalization pipeline in front of the comparison decides how much of the score is measurement. inference systems