• Corpus ID: 2973672

Machine learning with operational costs

@article{Tulabandhula2013MachineLW,
  title={Machine learning with operational costs},
  author={Theja Tulabandhula and Cynthia Rudin},
  journal={ArXiv},
  year={2013},
  volume={abs/1112.0698}
}
This work proposes a way to align statistical modeling with decision making. We provide a method that propagates the uncertainty in predictive modeling to the uncertainty in operational cost, where operational cost is the amount spent by the practitioner in solving the problem. The method allows us to explore the range of operational costs associated with the set of reasonable statistical models, so as to provide a useful way for practitioners to understand uncertainty. To do this, the… 

Figures and Tables from this paper

Robust Optimization using Machine Learning for Uncertainty Sets
TLDR
This work proposes a new uncertainty set design and shows how tools from statistical learning theory can be employed to provide probabilistic guarantees on the robustness of the policy.
On combining machine learning with decision making
TLDR
This work uses the MLOC framework to study a problem that has implications for power grid reliability and maintenance, called the Machine Learning and Traveling Repairman Problem (ML&TRP), and presents a new application and covering number bound for the framework.
Predictive and prescriptive methods in operations research and machine learning : an optimization approach
TLDR
This thesis proposes approaches that start from data leading to high-quality decisions and predictions in various application areas, and presents the method of prescriptive trees, that prescribes the best treatment option by learning from observational data while simultaneously predicting counterfactuals.
Decision-Driven Regularization: Harmonizing the Predictive and Prescriptive
TLDR
This work introduces the notion of a regularizer that measures the value of a predictive model in terms of the cost incurred in the decision process, and centred on the premise that the bias-variance trade-off in the learning problem is not transformed linearly by the subsequent decision problem.
A Blended Model for Predict-then-Optimize
TLDR
This paper critically shows that there are three perspectives, from which the predict-then-optimize problem may be approached, namely the regularization, the robust optimization, and the regret minimization approaches, and proves that these perspectives are equivalent to or can be naturally approximated to arrive at the proposed model.
Integrated Optimization of Predictive and Prescriptive Tasks
TLDR
A new framework directly integrating predictive tasks under prescriptive tasks in order to prescribe consistent decisions is proposed, and the parameters of predictive algorithm within a prescription problem are trained via bilevel optimization techniques.
Contextual Decision-making under Parametric Uncertainty and Data-driven Optimistic Optimization
TLDR
It is shown that the worst/best-case reward over the proposed parameter uncertainty set serves as a confidence bound on the reward by sizing the uncertainty set properly, and performance guarantees for robust contextual optimization in the offline setting are developed.
The Analytics of Robust Satisficing
TLDR
A new prescriptive analytics tool called robust satisficing is developed that uses data to help a satisficer achieve her target expected reward or consumption as well as possible under ambiguous risks and prediction uncertainty.
Iterative Value-Aware Model Learning
TLDR
A new model-based reinforcement learning (MBRL) framework that incorporates the underlying decision problem in learning the transition model of the environment, called Iterative VAML, that benefits from the structure of how the planning is performed (i.e., through approximate value iteration) to devise a simpler optimization problem.
Fast and Optimal Bayesian Approximations for Targeted Prediction
TLDR
This work designs a framework for constructing optimal, scalable, and simple approximations for targeted prediction under a Bayesian model, and introduces a procedure that identifies a set of near-optimal predictors.
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 74 REFERENCES
The Price of Robustness
TLDR
An approach is proposed that flexibly adjust the level of conservatism of the robust solutions in terms of probabilistic bounds of constraint violations, and an attractive aspect of this method is that the new robust formulation is also a linear optimization problem, so it naturally extend to discrete optimization problems in a tractable way.
A Soft Robust Model for Optimization Under Ambiguity
TLDR
A framework for robust optimization that relaxes the standard notion of robustness by allowing the decision maker to vary the protection level in a smooth way across the uncertainty set is proposed and connected closely to the theory of convex risk measures.
Machine Learning for the New York City Power Grid
TLDR
A general process for transforming historical electrical grid data into models that aim to predict the risk of failures for components and systems is introduced, and these models are sufficiently accurate to assist in maintaining New York City's electrical grid.
A framework for input uncertainty analysis
TLDR
To achieve this goal, metamodel-assisted bootstrapping is introduced, and its performance relative to other proposals for dealing with input uncertainty on two queueing examples is illustrated.
Theory and Applications of Robust Optimization
In this paper we survey the primary research, both theoretical and applied, in the area of robust optimization (RO). Our focus is on the computational attractiveness of RO approaches, as well as the
Regression Shrinkage and Selection via the Lasso
TLDR
A new method for estimation in linear models called the lasso, which minimizes the residual sum of squares subject to the sum of the absolute value of the coefficients being less than a constant, is proposed.
Workload forecasting for a call center: Methodology and a case study
TLDR
An arrival count model which is based on a mixed Poisson process approach is introduced which is applied to data from an Israeli Telecom company call center and shows that during most hours of the day the model can reach desired precision levels.
A process for predicting manhole events in Manhattan
We present a knowledge discovery and data mining process developed as part of the Columbia/Con Edison project on manhole event prediction. This process can assist with real-world prioritization
Mean-absolute deviation portfolio optimization model and its applications to Tokyo stock market
The purpose of this paper is to demonstrate that a portfolio optimization model using the L 1 risk (mean absolute deviation risk) function can remove most of the difficulties associated with the
Robust discrete optimization and network flows
TLDR
This work proposes a robust integer programming problem of moderately larger size that allows controlling the degree of conservatism of the solution in terms of probabilistic bounds on constraint violation, and proposes an algorithm for robust network flows that solves the robust counterpart by solving a polynomial number of nominal minimum cost flow problems in a modified network.
...
1
2
3
4
5
...