More Like this

1. Fast Sparse Decision Tree Optimization via Reference Ensembles

   https://doi.org/10.1609/aaai.v36i9.21194  

   McTavish, Hayden; Zhong, Chudi; Achermann, Reto; Karimalis, Ilias; Chen, Jacques; Rudin, Cynthia; Seltzer, Margo (June 2022, Proceedings of the AAAI Conference on Artificial Intelligence)

   Sparse decision tree optimization has been one of the most fundamental problems in AI since its inception and is a challenge at the core of interpretable machine learning. Sparse decision tree optimization is computationally hard, and despite steady effort since the 1960's, breakthroughs have been made on the problem only within the past few years, primarily on the problem of finding optimal sparse decision trees. However, current state-of-the-art algorithms often require impractical amounts of computation time and memory to find optimal or near-optimal trees for some real-world datasets, particularly those having several continuous-valued features. Given that the search spaces of these decision tree optimization problems are massive, can we practically hope to find a sparse decision tree that competes in accuracy with a black box machine learning model? We address this problem via smart guessing strategies that can be applied to any optimal branch-and-bound-based decision tree algorithm. The guesses come from knowledge gleaned from black box models. We show that by using these guesses, we can reduce the run time by multiple orders of magnitude while providing bounds on how far the resulting trees can deviate from the black box's accuracy and expressive power. Our approach enables guesses about how to bin continuous features, the size of the tree, and lower bounds on the error for the optimal decision tree. Our experiments show that in many cases we can rapidly construct sparse decision trees that match the accuracy of black box models. To summarize: when you are having trouble optimizing, just guess. 

   more » « less
2. Predicting Average Product Lifetime Using Physics-Based Gaussian Process Method in Early Design Stage

   https://doi.org/10.1115/DETC2020-22199  

   Wei, Xinpeng; Han, Daoru; Du, Xiaoping (August 2020, Proceedings of the ASME 2020 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference)

   null (Ed.)

   Abstract Average lifetime, or mean time to failure (MTTF), of a product is an important metric to measure the product reliability. Current methods of evaluating MTTF are mainly statistics or data based. They need lifetime testing on a number of products to get the lifetime samples, which are then used to estimate MTTF. The lifetime testing, however, is expensive in terms of both time and cost. The efficiency is also low because it cannot be effectively incorporated in the early design stage where many physics-based models are available. We propose to predict MTTF in the design stage by means of physics-based models. The advantage is that the design can be continually improved by changing design variables until reliability measures, including MTTF, are satisfied. Since the physics-based models are usually computationally demanding, we face a problem with both big data (on the model input side) and small data (on the model output side). We develop an adaptive supervised training method based on Gaussian process regression, and the method can then quickly predict MTTF with minimized number of calling the physics-based models. The effectiveness of the method is demonstrated by two examples. 

   more » « less
3. Physics-Based Gaussian Process Method for Predicting Average Product Lifetime in Design Stage

   https://doi.org/10.1115/1.4049509  

   Wei, Xinpeng; Han, Daoru; Du, Xiaoping (August 2021, Journal of Computing and Information Science in Engineering)

   null (Ed.)

   Abstract The average lifetime or the mean time to failure (MTTF) of a product is an important metric to measure the product reliability. Current methods of evaluating the MTTF are mainly based on statistics or data. They need lifetime testing on a number of products to get the lifetime samples, which are then used to estimate the MTTF. The lifetime testing, however, is expensive in terms of both time and cost. The efficiency is also low because it cannot be effectively incorporated in the early design stage where many physics-based models are available. We propose to predict the MTTF in the design stage by means of a physics-based Gaussian process (GP) method. Since the physics-based models are usually computationally demanding, we face a problem with both big data (on the model input side) and small data (on the model output side). The proposed adaptive supervised training method with the Gaussian process regression can quickly predict the MTTF with a reduced number of physical model calls. The proposed method can enable continually improved design by changing design variables until reliability measures, including the MTTF, are satisfied. The effectiveness of the method is demonstrated by three examples. 

   more » « less
4. Decentralized sequential active hypothesis testing and the MAC feedback capacity

   Anastasopoulos, Achilleas and (June 2020, Proceedings of IEEE International Symposium on Information Theory)

   We consider the problem of decentralized sequential active hypothesis testing (DSAHT), where two transmitting agents, each possessing a private message, are actively helping a third agent–and each other–to learn the message pair over a discrete memoryless multiple access channel (DM-MAC). The third agent (receiver) observes the noisy channel output, which is also available to the transmitting agents via noiseless feedback. We formulate this problem as a decentralized dynamic team, show that optimal transmission policies have a time-invariant domain, and characterize the solution through a dynamic program. Several alternative formulations are discussed involving time-homogenous cost functions and/or variable-length codes, resulting in solutions described through fixed-point, Bellman-type equations. Subsequently, we make connections with the problem of simplifying the multi-letter capacity expressions for the noiseless feedback capacity of the DM-MAC. We show that restricting attention to distributions induced by optimal transmission schemes for the DSAHT problem, without loss of optimality, transforms the capacity expression, so that it can be thought of as the average reward received by an appropriately defined stochastic dynamical system with time-invariant state space. 

   more » « less
5. Pretrial release judgments and decision fatigue

   Shroff, Ravi; Vamvourellis, Konstantinos (November 2022, Judgment and decision making)

   Field studies in many domains have found evidence of decision fatigue, a phenomenon describing how decision quality can be impaired by the act of making previous decisions. Debate remains, however, over posited psychological mechanisms underlying decision fatigue, and the size of effects in high-stakes settings. We examine an extensive set of pretrial arraignments in a large, urban court system to investigate how judicial release and bail decisions are influenced by the time an arraignment occurs. We find that release rates decline modestly in the hours before lunch and before dinner, and these declines persist after statistically adjusting for an extensive set of observed covariates. However, we find no evidence that arraignment time affects pretrial release rates in the remainder of each decision-making session. Moreover, we find that release rates remain unchanged after a meal break even though judges have the opportunity to replenish their mental and physical resources by resting and eating. In a complementary analysis, we find that the rate at which judges concur with prosecutorial bail requests does not appear to be influenced by either arraignment time or a meal break. Taken together, our results imply that to the extent that decision fatigue plays a role in pretrial release judgments, effects are small and inconsistent with previous explanations implicating psychological depletion processes. 

   more » « less