Combining Strategies Efficiently: High-Quality Decisions from Conflicting Advice
Wouter M. Koolen
Abstract:
Combining Strategies Efficiently: High-Quality Decisions from Conflicting Advice
Wouter M. Koolen
In this dissertation we study machine learning: the automated
discovery and exploitation of regularities in data. We may use
regularities identified in objects to explain the past
(e.g. archaeology, justice), as well as regularities found in
processes to predict the future (e.g. weather, stock market) and guide
our actions.
With ubiquitous computational resources, machine learning algorithms
have become pervasive. For example, they manage financial portfolios
and power-saving policy, provide personalised movie recommendations as
well as advertisements, and form the core of state-of-the-art data
compression software.
This dissertation develops the theory of online learning, a branch of
machine learning that investigates sequential decision problems with
immediate feedback. In particular, we study the setting called
prediction with expert advice. Our task is to predict a sequence of
data. Each trial, we may first consult a given set of experts. We then
combine their advice and issue our prediction of the next
outcome. Finally, the next outcome is revealed, and we incur loss
based on the discrepancy between our prediction and it.
The goal is to build efficient algorithms with small regret, i.e. the
difference between the incurred cumulative loss and the loss of the
best strategy in hindsight from a fixed reference class. In this
sense, the strategies in the reference class are the patterns, and
achieving small regret means learning which reference strategy best
models the data. The main difference between the learning problems we
consider is the complexity of the reference set. Algorithms for
prediction with expert advice have many applications including
classification, regression, hypothesis testing, model selection, data
compression, gambling and investing in the stock market.
In Chapter 2 we give a game-theoretic analysis of the simplest online
learning problem, the prediction of a sequence of binary outcomes
under 0/1 loss with the help of two experts. For this simple problem,
we compute the minimax, i.e. game-theoretically optimal, regret, and
show how to implement the optimal strategy efficiently. We then give
special attention to the case that one of the experts is good. We
conclude with a new result: the optimal algorithm for competing with
the set of meta-experts that switch between the two basic experts.
In Chapter 3 we show how models for prediction with expert advice can
be defined concisely and clearly using hidden Markov models (HMMs);
standard algorithms can then be used to efficiently calculate how the
expert predictions should be weighted. We focus on algorithms for
tracking the best expert. Here the strategies in the reference set
follow the advice of a single expert, but this expert may change
between trials. We cast existing models as HMMs, starting from the
fixed share algorithm, recover the running times and regret bounds for
each algorithm, and discuss how they are related. We also describe
three new models for switching between experts.
In Chapter 4 we extend the setting to tracking the best learning
expert. Whereas vanilla experts can be tapped for advice about the
current trial, learning experts may be queried for advice given each
possible subset of the past data. This additional power is available
to both the algorithm and the reference strategies. Achieving small
regret thus means learning how to partition the trials, and which
learning expert to train and follow within each partition cell. We
give efficient algorithms with small regret for tracking learning
experts that can themselves be formalised using the expert HMMs of
Chapter 3.
In Chapter 5 we consider reference strategies that switch between two
experts based on their cumulative loss instead of on time. This
chapter is formulated in financial terms to make the presentation more
intuitive. We present a simple online two-way trading algorithm that
exploits fluctuations in the unit price of an asset. Rather than
analysing worst-case performance under some assumptions, we prove a
novel, unconditional performance bound that is parameterised either by
the actual dynamics of the price of the asset, or by a simplifying
model thereof. We discuss application of the results to prediction
with expert advice, data compression and hypothesis testing.
In Chapter 6 we consider prediction with structured concepts. Each
round we select a concept, which is composed of components. The loss
of a concept is the sum of the losses of its components. Whereas the
losses of different components are independent, the losses of
different concepts are highly related. We develop an online algorithm,
called Component Hedge that exploits this dependence, and thereby
avoids the so called range factor that arises when the dependences are
ignored. We show that Component Hedge has optimal regret bounds for a
large variety of structured concept classes.