Alert button
Picture for Xinyi Chen

Xinyi Chen

Alert button

Princeton University

Machine Learning for Mechanical Ventilation Control

Feb 26, 2021
Daniel Suo, Cyril Zhang, Paula Gradu, Udaya Ghai, Xinyi Chen, Edgar Minasyan, Naman Agarwal, Karan Singh, Julienne LaChance, Tom Zajdel, Manuel Schottdorf, Daniel Cohen, Elad Hazan

Figure 1 for Machine Learning for Mechanical Ventilation Control
Figure 2 for Machine Learning for Mechanical Ventilation Control
Figure 3 for Machine Learning for Mechanical Ventilation Control
Figure 4 for Machine Learning for Mechanical Ventilation Control

We consider the problem of controlling an invasive mechanical ventilator for pressure-controlled ventilation: a controller must let air in and out of a sedated patient's lungs according to a trajectory of airway pressures specified by a clinician. Hand-tuned PID controllers and similar variants have comprised the industry standard for decades, yet can behave poorly by over- or under-shooting their target or oscillating rapidly. We consider a data-driven machine learning approach: First, we train a simulator based on data we collect from an artificial lung. Then, we train deep neural network controllers on these simulators.We show that our controllers are able to track target pressure waveforms significantly better than PID controllers. We further show that a learned controller generalizes across lungs with varying characteristics much more readily than PID controllers do.

Viaarxiv icon

FUEL: Fast UAV Exploration using Incremental Frontier Structure and Hierarchical Planning

Oct 22, 2020
Boyu Zhou, Yichen Zhang, Xinyi Chen, Shaojie Shen

Figure 1 for FUEL: Fast UAV Exploration using Incremental Frontier Structure and Hierarchical Planning
Figure 2 for FUEL: Fast UAV Exploration using Incremental Frontier Structure and Hierarchical Planning
Figure 3 for FUEL: Fast UAV Exploration using Incremental Frontier Structure and Hierarchical Planning
Figure 4 for FUEL: Fast UAV Exploration using Incremental Frontier Structure and Hierarchical Planning

Autonomous exploration is a fundamental problem for various applications of unmanned aerial vehicles. Existing methods, however, were demonstrated to have low efficiency, due to the lack of optimality consideration, conservative motion plans and low decision frequencies. In this paper, we propose FUEL, a hierarchical framework that can support Fast UAV Exploration in complex unknown environments. We maintain crucial information in the entire space required by exploration planning by a frontier information structure (FIS), which can be updated incrementally when the space is explored. Supported by the FIS, a hierarchical planner plan exploration motions in three steps, which find efficient global coverage paths, refine a local set of viewpoints and generate minimum-time trajectories in sequence. We present extensive benchmark and real-world tests, in which our method completes the exploration tasks with unprecedented efficiency (3-8 times faster) compared to state-of-the-art approaches. Our method will be made open source to benefit the community.

* Video: https://www.youtube.com/watch?v=_dGgZUrWk-8; Demo: https://github.com/HKUST-Aerial-Robotics/FUEL 
Viaarxiv icon

Black-Box Control for Linear Dynamical Systems

Jul 13, 2020
Xinyi Chen, Elad Hazan

We consider the problem of controlling an unknown linear time-invariant dynamical system from a single chain of black-box interactions, and with no access to resets or offline simulation. Under the assumption that the system is controllable, we give the first efficient algorithm that is capable of attaining sublinear regret in a single trajectory under the setting of online nonstochastic control. We give finite-time regret bound of our algorithm, as well as a nearly-matching lower bound that shows this regret to be almost best-attainable by any algorithm.

Viaarxiv icon

Online Agnostic Boosting via Regret Minimization

Mar 02, 2020
Nataly Brukhim, Xinyi Chen, Elad Hazan, Shay Moran

Figure 1 for Online Agnostic Boosting via Regret Minimization

Boosting is a widely used machine learning approach based on the idea of aggregating weak learning rules. While in statistical learning numerous boosting methods exist both in the realizable and agnostic settings, in online learning they exist only in the realizable case. In this work we provide the first agnostic online boosting algorithm; that is, given a weak learner with only marginally-better-than-trivial regret guarantees, our algorithm boosts it to a strong learner with sublinear regret. Our algorithm is based on an abstract (and simple) reduction to online convex optimization, which efficiently converts an arbitrary online convex optimizer to an online booster. Moreover, this reduction extends to the statistical as well as the online realizable settings, thus unifying the 4 cases of statistical/online and agnostic/realizable boosting.

Viaarxiv icon

Calibration, Entropy Rates, and Memory in Language Models

Jun 11, 2019
Mark Braverman, Xinyi Chen, Sham M. Kakade, Karthik Narasimhan, Cyril Zhang, Yi Zhang

Figure 1 for Calibration, Entropy Rates, and Memory in Language Models
Figure 2 for Calibration, Entropy Rates, and Memory in Language Models
Figure 3 for Calibration, Entropy Rates, and Memory in Language Models
Figure 4 for Calibration, Entropy Rates, and Memory in Language Models

Building accurate language models that capture meaningful long-term dependencies is a core challenge in natural language processing. Towards this end, we present a calibration-based approach to measure long-term discrepancies between a generative sequence model and the true distribution, and use these discrepancies to improve the model. Empirically, we show that state-of-the-art language models, including LSTMs and Transformers, are \emph{miscalibrated}: the entropy rates of their generations drift dramatically upward over time. We then provide provable methods to mitigate this phenomenon. Furthermore, we show how this calibration-based approach can also be used to measure the amount of memory that language models use for prediction.

Viaarxiv icon

Extreme Tensoring for Low-Memory Preconditioning

Feb 12, 2019
Xinyi Chen, Naman Agarwal, Elad Hazan, Cyril Zhang, Yi Zhang

Figure 1 for Extreme Tensoring for Low-Memory Preconditioning
Figure 2 for Extreme Tensoring for Low-Memory Preconditioning
Figure 3 for Extreme Tensoring for Low-Memory Preconditioning
Figure 4 for Extreme Tensoring for Low-Memory Preconditioning

State-of-the-art models are now trained with billions of parameters, reaching hardware limits in terms of memory consumption. This has created a recent demand for memory-efficient optimizers. To this end, we investigate the limits and performance tradeoffs of memory-efficient adaptively preconditioned gradient methods. We propose extreme tensoring for high-dimensional stochastic optimization, showing that an optimizer needs very little memory to benefit from adaptive preconditioning. Our technique applies to arbitrary models (not necessarily with tensor-shaped parameters), and is accompanied by regret and convergence guarantees, which shed light on the tradeoffs between preconditioner quality and expressivity. On a large-scale NLP model, we reduce the optimizer memory overhead by three orders of magnitude, without degrading performance.

Viaarxiv icon

Online Learning of Quantum States

Oct 01, 2018
Scott Aaronson, Xinyi Chen, Elad Hazan, Satyen Kale, Ashwin Nayak

Suppose we have many copies of an unknown $n$-qubit state $\rho$. We measure some copies of $\rho$ using a known two-outcome measurement $E_{1}$, then other copies using a measurement $E_{2}$, and so on. At each stage $t$, we generate a current hypothesis $\sigma_{t}$ about the state $\rho$, using the outcomes of the previous measurements. We show that it is possible to do this in a way that guarantees that $|\operatorname{Tr}(E_{i} \sigma_{t}) - \operatorname{Tr}(E_{i}\rho) |$, the error in our prediction for the next measurement, is at least $\varepsilon$ at most $\operatorname{O}\!\left(n / \varepsilon^2 \right) $ times. Even in the "non-realizable" setting---where there could be arbitrary noise in the measurement outcomes---we show how to output hypothesis states that do significantly worse than the best possible states at most $\operatorname{O}\!\left(\sqrt {Tn}\right) $ times on the first $T$ measurements. These results generalize a 2007 theorem by Aaronson on the PAC-learnability of quantum states, to the online and regret-minimization settings. We give three different ways to prove our results---using convex optimization, quantum postselection, and sequential fat-shattering dimension---which have different advantages in terms of parameters and portability.

* 17 pages 
Viaarxiv icon

The Case for Full-Matrix Adaptive Regularization

Jun 08, 2018
Naman Agarwal, Brian Bullins, Xinyi Chen, Elad Hazan, Karan Singh, Cyril Zhang, Yi Zhang

Figure 1 for The Case for Full-Matrix Adaptive Regularization
Figure 2 for The Case for Full-Matrix Adaptive Regularization
Figure 3 for The Case for Full-Matrix Adaptive Regularization
Figure 4 for The Case for Full-Matrix Adaptive Regularization

Adaptive regularization methods come in diagonal and full-matrix variants. However, only the former have enjoyed widespread adoption in training large-scale deep models. This is due to the computational overhead of manipulating a full matrix in high dimension. In this paper, we show how to make full-matrix adaptive regularization practical and useful. We present GGT, a truly scalable full-matrix adaptive optimizer. At the heart of our algorithm is an efficient method for computing the inverse square root of a low-rank matrix. We show that GGT converges to first-order local minima, providing the first rigorous theoretical analysis of adaptive regularization in non-convex optimization. In preliminary experiments, GGT trains faster across a variety of synthetic tasks and standard deep learning benchmarks.

Viaarxiv icon