Top Qs
Timeline
Chat
Perspective

Competitive regret

Concept in decision theory From Wikipedia, the free encyclopedia

Remove ads

In decision theory and machine learning, competitive regret refers to a performance measure that evaluates an algorithm's regret relative to an oracle or benchmark strategy. Unlike traditional regret, which compares against the best fixed decision in hindsight, competitive regret compares against decision-makers with different capabilities—either with greater computational resources or access to additional information.

The formal definition of competitive regret typically involves a ratio or difference between the regret of an algorithm and the regret of a reference oracle. An algorithm is considered to have "good" competitive regret if this ratio remains bounded even as the problem size increases.

This framework has applications in various domains including online optimization, reinforcement learning, portfolio selection, and multi-armed bandit problems. Competitive regret analysis provides researchers with a more nuanced evaluation metric than standard regret, helping them develop algorithms that can achieve near-optimal performance even under practical constraints and uncertainty.

Remove ads

Competitive regret to the oracle with full power

Consider estimating a discrete probability distribution on a discrete set based on data , the regret of an estimator[1] is defined as

where is the set of all possible probability distribution, and

where is the Kullback–Leibler divergence between and .

Remove ads

Competitive regret to the oracle with limited power

Summarize
Perspective

Oracle with partial information

The oracle is restricted to have access to partial information of the true distribution by knowing the location of in the parameter space up to a partition.[1] Given a partition of the parameter space, and suppose the oracle knows the subset where the true . The oracle will have regret as

The competitive regret to the oracle will be

Oracle with partial information

The oracle knows exactly , but can only choose the estimator among natural estimators. A natural estimator assigns equal probability to the symbols which appear the same number of time in the sample.[1] The regret of the oracle is

and the competitive regret is

Remove ads

Example

For the estimator proposed in Acharya et al.(2013),[2]

Here denotes the k-dimensional unit simplex surface. The partition denotes the permutation class on , where and are partitioned into the same subset if and only if is a permutation of .

Remove ads

References

Loading related searches...

Wikiwand - on

Seamless Wikipedia browsing. On steroids.

Remove ads