Task-based End-to-end Model Learning in Stochastic Optimization

By Priya Lekha Donti, J. Kolter, Brandon Amos. Neural Information Processing Systems, 2017

462 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/7b581c9ce200b031451f592478c7c34b5fc47898

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Differentiable Convex Optimization Layers (2019, background): “These layers have found several applications [40, 6, 35, 27, 5, 53, 75, 52, 12, 11], but many applications remain relatively unexplored (see, e.g. , [4, §8]).”
Tackling Climate Change with Machine Learning (2019, background): “forecasts since weather distributions shift over time [57]. In addition to incorporating system physics, ML models should also directly optimize for system goals [58–60]. For instance, the authors of [58] use a deep neural network to produce demand forecasts that optimize for electricity scheduling ”
Strong mixed-integer programming formulations for trained neural networks (2018, background): “Broadly, our work fits into a growing body of research in prescriptive analytics and specifically the “predict, then optimize” framework, which considers how to embed trained machine learning models into optimization problems [11,12,17,18,24,28,39].”
Universal Planning Networks (2018, background): “Amos & Kolter (2017); Donti et al. (2017) also look at embedding differentiable optimization procedures (quadratic programs) within neural networks.”
SATNet: Bridging deep learning and logical reasoning using a differentiable satisfiability solver (2019, background): “For instance, previous work has introduced differentiable modules for quadratic programs (Amos & Kolter, 2017; Donti et al., 2017), submodular optimization problems (Djolonga & Krause, 2017; Tschiatschek et al., 2018; Wilder et al., 2018), and equilibrium computation in zero-sum games (Ling et al.,…”
Decision Trees for Decision-Making under the Predict-then-Optimize Framework (2020, unknown): “Donti et al. (2017) provide a more general methodology related to this line of work that relies on diﬀerentiating the optimization problem.”
Melding the Data-Decisions Pipeline: Decision-Focused Learning for Combinatorial Optimization (2018, methodology): “Similar ideas have recently been explored in the context of convex optimization (Donti, Amos, and Kolter 2017), but to our knowledge ours is the ﬁrst attempt to train machine learning systems for performance on combinatorial decision-making problems.”
Iterative Value-Aware Model Learning (2018, background): “For example, it might be the coefficients of the cost function [Elmachtoub and Grigas, 2017, Kao and Van Roy, 2014] or the probability distribution in a stochastic program [Donti et al., 2017].”
Lagrangian Duality for Constrained Deep Learning (2020, background): “e, but is receiving increasing attention, such as the efforts in jointly training prediction and optimization models [38, 25, 27] and incorporating optimization algorithms into differentiable systems [17, 4, 39]. This research originated in an attempt to apply deep learning to fundamentally differen”
Implicit Deep Learning (2019, background): “nforcement learning, which can be seen as a kind of implicit architecture. In [2] the authors introduced implicit layers where the activation is the solution of some quadratic programming problem; in [18], the authors incorporate stochastic optimization formulation for end-to-end learning task, in w”
Smart "Predict, then Optimize" (2017, methodology): “Donti et al. (2017) proposes a heuristic to address a more general setting than that of Kao et al. (2009), and also focus on the case of quadratic optimization.”