Power-law relationship
A power-law relationship is a mathematical relationship in which one quantity is proportional to another quantity raised to a fixed exponent.
Timeless Algorithms, The Foundational Ideas
A power-law relationship is one in which a quantity varies as a power of another quantity.
- How it is used
- Summarizes how language-model loss changes with parameters, training data, and compute, supporting estimates for larger training runs.
- How deep it goes
- Derives the effect of proportional resource increases from a simplified equation and interprets log-log plots, while warning that apparent straightness is inconclusive.
- Examples
- Uses word frequencies to contrast distributions with scaling and calculates loss at four resource levels under an exponent of −0.1.
- What you can do after
- Evaluate power-law scaling claims by interpreting proportional changes, resource constraints, and the limits of log-log evidence.
Read how Timeless Algorithms, The Foundational Ideas explains power-law relationship →
Statistics Slam Dunk
A power law is a mathematical relationship in which one quantity varies as a constant power of another quantity.
- How it is used
- Frames uneven NBA outcomes and apparent success streaks as questions about nonlinear relationships, concentration, randomness, and causality.
- How deep it goes
- Defines proportional power relationships and applies them to Pareto distributions, stopping short of formal derivation.
- Examples
- Uses NBA franchise longevity, games played and won, and the hot-hand success sequence as concrete applications.
- What you can do after
- Evaluate/understand right-skewed outcomes, Pareto concentration, and claims that streaks reflect causality rather than randomness.
Read how Statistics Slam Dunk explains power-law relationship →
Related concepts
- linearitycontrasts with · Linearity is the property of a relationship, function, or transformation being governed by a linear rule, such as proportional change or preservation of linear structure.
- power law distributioncontrasts with · A power-law distribution is a probability distribution in which the frequency or tail probability of values varies as a power of their size.
- regression to the meancontrasts with · Regression to the mean is the tendency for unusually high or low measurements to be followed by measurements closer to the underlying population average.
- scaling lawused by · A scaling law is an empirical relationship describing how a system’s properties change as its scale or resource levels vary.
- randomnesscontrasts with · Randomness is the property of a process or outcome involving chance or probabilistic variation rather than being fully determined by available conditions.
Questions about power-law relationship
What is power-law relationship?
A power-law relationship is a mathematical relationship in which one quantity is proportional to another quantity raised to a fixed exponent.
Which Manning books explain power-law relationship?
Timeless Algorithms, The Foundational Ideas and Statistics Slam Dunk.
What is the difference between power-law relationship and linearity?
A power-law relationship is a mathematical relationship in which one quantity is proportional to another quantity raised to a fixed exponent. Linearity is the property of a relationship, function, or transformation being governed by a linear rule, such as proportional change or preservation of linear structure.
What kinds of power-law relationship are there?
Kinds of power-law relationship include Pareto distribution.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.

