You need to log in to edit.

You can create a new account if you don't have one.

Or, discuss a change on Slack.

You can create a new account if you don't have one.

Or, discuss a change on Slack.

1 code implementation • 14 Feb 2024 • Harrison Delecki, Marcell Vazquez-Chanlatte, Esen Yel, Kyle Wray, Tomer Arnon, Stefan Witwicki, Mykel J. Kochenderfer

However, model-based planners may be brittle under these types of uncertainty because they rely on an exact model and tend to commit to a single optimal behavior.

1 code implementation • 29 Jan 2024 • Alexandros E. Tzikas, Licio Romao, Mert Pilanci, Alessandro Abate, Mykel J. Kochenderfer

Many machine learning applications require operating on a spatially distributed dataset.

no code implementations • 22 Jan 2024 • Jiachen Li, Chuanbo Hua, Hengbo Ma, Jinkyoo Park, Victoria Dax, Mykel J. Kochenderfer

In this paper, we propose a systematic relational reasoning approach with explicit inference of the underlying dynamically evolving relational structures, and we demonstrate its effectiveness for multi-agent trajectory prediction and social robot navigation.

no code implementations • 18 Jan 2024 • Ali Baheri, Mykel J. Kochenderfer

This paper explores the integration of optimal transport (OT) theory with multi-agent reinforcement learning (MARL).

no code implementations • 11 Jan 2024 • Victoria M. Dax, Jiachen Li, Kevin Leahy, Mykel J. Kochenderfer

Graph-structured data is ubiquitous throughout natural and social sciences, and Graph Neural Networks (GNNs) have recently been shown to be effective at solving prediction and inference problems on graph data.

no code implementations • 7 Jan 2024 • Victoria M. Dax, Jiachen Li, Enna Sachdeva, Nakul Agarwal, Mykel J. Kochenderfer

The results show superior performance compared to existing methods in modeling spatio-temporal relations, motion prediction, and identifying time-invariant latent features.

no code implementations • 27 Nov 2023 • Jiachen Li, David Isele, Kanghoon Lee, Jinkyoo Park, Kikuo Fujimura, Mykel J. Kochenderfer

Moreover, we propose an interactivity estimation mechanism based on the difference between predicted trajectories in these two situations, which indicates the degree of influence of the ego agent on other agents.

1 code implementation • 31 Oct 2023 • Kyle Brown, Dylan M. Asmar, Mac Schwager, Mykel J. Kochenderfer

Mobile autonomous robots have the potential to revolutionize manufacturing processes.

no code implementations • 30 Oct 2023 • Arec Jamgochian, Hugo Buurmeijer, Kyle H. Wray, Anthony Corso, Mykel J. Kochenderfer

Optimal plans in Constrained Partially Observable Markov Decision Processes (CPOMDPs) maximize reward objectives while satisfying hard cost constraints, generalizing safe planning under state and transition uncertainty.

1 code implementation • 25 Sep 2023 • Bernard Lange, Jiachen Li, Mykel J. Kochenderfer

We introduce the Scene Informer, a unified approach for predicting both observed agent trajectories and inferring occlusions in a partially observable setting.

no code implementations • 21 Sep 2023 • Marc R. Schlichting, Nina V. Boord, Anthony L. Corso, Mykel J. Kochenderfer

The validation can be accelerated by identifying critical failure scenarios for the system under test and by reducing the simulation runtime.

no code implementations • 20 Jul 2023 • Anthony Corso, David Karamadian, Romeo Valentin, Mary Cooper, Mykel J. Kochenderfer

As machine learning (ML) systems increasingly permeate high-stakes settings such as healthcare, transportation, military, and national security, concerns regarding their reliability have emerged.

no code implementations • 19 Jul 2023 • Kanghoon Lee, Jiachen Li, David Isele, Jinkyoo Park, Kikuo Fujimura, Mykel J. Kochenderfer

Although deep reinforcement learning (DRL) has shown promising results for autonomous navigation in interactive traffic scenarios, existing work typically adopts a fixed behavior policy to control social vehicles in the training environment.

no code implementations • 3 Jul 2023 • Sydney M. Katz, Anthony L. Corso, Esen Yel, Mykel J. Kochenderfer

Perception systems operate as a subcomponent of the general autonomy stack, and perception system designers often need to optimize performance characteristics while maintaining safety with respect to the overall closed-loop system.

1 code implementation • 19 Jun 2023 • Elysia Q. Smyers, Sydney M. Katz, Anthony L. Corso, Mykel J. Kochenderfer

We also provide an interface that evaluates trained models on slices of this dataset to identify changes in performance with respect to changing environmental conditions.

no code implementations • 31 May 2023 • Robert J. Moss, Anthony Corso, Jef Caers, Mykel J. Kochenderfer

BetaZero learns offline approximations that replace heuristics to enable online decision making in long-horizon problems.

1 code implementation • 17 May 2023 • Harrison Delecki, Anthony Corso, Mykel J. Kochenderfer

Estimating the distribution over failures is a key step in validating autonomous systems.

no code implementations • 10 May 2023 • Ali Baheri, Mykel J. Kochenderfer

We propose a joint falsification and fidelity optimization framework for safety validation of autonomous systems.

1 code implementation • 3 May 2023 • Robert J. Moss, Mykel J. Kochenderfer, Maxime Gariel, Arthur Dubois

Accurately estimating the probability of failure for safety-critical systems is important for certification.

1 code implementation • 19 Apr 2023 • Yizheng Wang, Markus Zechner, Gege Wen, Anthony Louis Corso, John Michael Mern, Mykel J. Kochenderfer, Jef Karel Caers

In this study, we address these issues by modeling the decision-making process for carbon storage operations as a partially observable Markov decision process (POMDP).

1 code implementation • 17 Mar 2023 • Soyeon Jung, Mykel J. Kochenderfer

For each segment, a Gaussian mixture model is used to learn the deviations of aircraft trajectories from their procedures.

1 code implementation • 28 Dec 2022 • Zahra Shahrooei, Mykel J. Kochenderfer, Ali Baheri

Simulation-based falsification is a practical testing method to increase confidence that the system will meet safety requirements.

1 code implementation • 23 Dec 2022 • Arec Jamgochian, Anthony Corso, Mykel J. Kochenderfer

Rather than augmenting rewards with penalties for undesired behavior, Constrained Partially Observable Markov Decision Processes (CPOMDPs) plan safely by imposing inviolable hard constraint value budgets.

no code implementations • 22 Nov 2022 • Anthony Corso, Kyu-Young Kim, Shubh Gupta, Grace Gao, Mykel J. Kochenderfer

An important step in the design of autonomous systems is to evaluate the probability that a failure will occur.

1 code implementation • 16 Nov 2022 • Masha Itkina, Mykel J. Kochenderfer

We propose the use of evidential deep learning to estimate the epistemic uncertainty over a low-dimensional, interpretable latent space in a trajectory prediction setting.

1 code implementation • 31 Oct 2022 • Jennifer She, Jayesh K. Gupta, Mykel J. Kochenderfer

Sparse and delayed rewards pose a challenge to single agent reinforcement learning.

Multi-agent Reinforcement Learning
reinforcement-learning
**+1**

no code implementations • 25 Oct 2022 • Anthony Corso, Yizheng Wang, Markus Zechner, Jef Caers, Mykel J. Kochenderfer

This POMDP model can be used as a test bed to drive the development of novel decision-making algorithms for CCS operations.

1 code implementation • 10 Oct 2022 • Michael H. Lim, Tyler J. Becker, Mykel J. Kochenderfer, Claire J. Tomlin, Zachary N. Sunberg

Thus, when combined with sparse sampling MDP algorithms, this approach can yield algorithms for POMDPs that have no direct theoretical dependence on the size of the state and observation spaces.

1 code implementation • 3 Oct 2022 • Bernard Lange, Masha Itkina, Mykel J. Kochenderfer

Environment prediction frameworks are integral for autonomous vehicles, enabling safe navigation in dynamic environments.

1 code implementation • 30 Sep 2022 • Lisa J. Einstein, Robert J. Moss, Mykel J. Kochenderfer

However, decision makers struggle to determine optimal evacuation policies given the chaos, uncertainty, and value judgments inherent in emergency evacuations.

no code implementations • 28 Sep 2022 • Nicholas Rober, Sydney M. Katz, Chelsea Sidrane, Esen Yel, Michael Everett, Mykel J. Kochenderfer, Jonathan P. How

As neural networks (NNs) become more prevalent in safety-critical applications such as control of vehicles, there is a growing need to certify that systems with NN components are safe.

1 code implementation • 27 Sep 2022 • Dylan M. Asmar, Mykel J. Kochenderfer

The assumption of a collaborative environment enables us to use the agent's policy to estimate the distribution over action suggestions.

no code implementations • 27 Sep 2022 • Maneekwan Toyungyernsub, Esen Yel, Jiachen Li, Mykel J. Kochenderfer

Detection and segmentation of moving obstacles, along with prediction of the future occupancy states of the local environment, are essential for autonomous vehicles to proactively make safe and informed decisions.

1 code implementation • 16 Sep 2022 • Joshua Ott, Edward Balaban, Mykel J. Kochenderfer

Adaptive Informative Path Planning with Multimodal Sensing (AIPPMS) considers the problem of an agent equipped with multiple sensors, each with different sensing accuracy and energy costs.

no code implementations • 15 Sep 2022 • Kyle Hollins Wray, Stas Tiomkin, Mykel J. Kochenderfer, Pieter Abbeel

Multi-objective optimization models that encode ordered sequential constraints provide a solution to model various challenging problems including encoding preferences, modeling a curriculum, and enforcing measures of safety.

no code implementations • 12 Sep 2022 • Joshua Ott, Sung-Kyun Kim, Amanda Bouman, Oriana Peltzer, Mamoru Sobue, Harrison Delecki, Mykel J. Kochenderfer, Joel Burdick, Ali-akbar Agha-mohammadi

Robotic exploration of unknown environments is fundamentally a problem of decision making under uncertainty where the robot must account for uncertainty in sensor measurements, localization, action execution, as well as many other factors.

no code implementations • 10 Aug 2022 • Jiachen Li, Chuanbo Hua, Jinkyoo Park, Hengbo Ma, Victoria Dax, Mykel J. Kochenderfer

While the modeling of pair-wise relations has been widely studied in multi-agent interacting systems, its ability to capture higher-level and larger-scale group-wise activities is limited.

no code implementations • 1 Jun 2022 • Junyoung Park, Federico Berto, Arec Jamgochian, Mykel J. Kochenderfer, Jinkyoo Park

In this paper, we propose Meta-SysId, a meta-learning approach to model sets of systems that have behavior governed by common but unknown laws and that differentiate themselves by their context.

1 code implementation • 21 May 2022 • Anthony L. Corso, Sydney M. Katz, Craig Innes, Xin Du, Subramanian Ramamoorthy, Mykel J. Kochenderfer

We formulate a risk function to quantify the effect of a given perceptual error on overall safety, and show how we can use it to design safer perception systems by including a risk-dependent term in the loss function and generating training data in risk-sensitive regions.

1 code implementation • 30 Mar 2022 • Dylan M. Asmar, Ransalu Senanayake, Shawn Manuel, Mykel J. Kochenderfer

We generalize the derivation of model predictive path integral control (MPPI) to allow for a single joint distribution across controls in the control sequence.

1 code implementation • 26 Mar 2022 • Harrison Delecki, Masha Itkina, Bernard Lange, Ransalu Senanayake, Mykel J. Kochenderfer

This paper presents a method for characterizing failures of LiDAR-based perception systems for AVs in adverse weather conditions.

no code implementations • 11 Mar 2022 • Christopher Lazarus, Mykel J. Kochenderfer

We compare the runtime of our approach against state-of-the-art verification algorithms for full-precision neural networks.

no code implementations • 11 Mar 2022 • Christopher Lazarus, Mykel J. Kochenderfer

We present an RL algorithm tailored specifically for BNNs.

no code implementations • 6 Mar 2022 • Xiaobai Ma, David Isele, Jayesh K. Gupta, Kikuo Fujimura, Mykel J. Kochenderfer

Multi-agent reinforcement learning (MARL) provides an efficient way for simultaneously learning policies for multiple agents interacting with each other.

Multi-agent Reinforcement Learning
reinforcement-learning
**+1**

no code implementations • 4 Feb 2022 • Chelsea Sidrane, Sydney Katz, Anthony Corso, Mykel J. Kochenderfer

When the forward model that produced the observations is nonlinear and stochastic, solving the inverse problem is very challenging.

1 code implementation • 8 Jan 2022 • Arec Jamgochian, Di wu, Kunal Menda, Soyeon Jung, Mykel J. Kochenderfer

In this paper, we introduce the conditional approximate normalizing flow (CANF) to make probabilistic multi-step time-series forecasts when correlations are present over long time horizons.

1 code implementation • 7 Dec 2021 • Liam A. Kruse, Allan L. Reiss, Mykel J. Kochenderfer, Stephanie Balters

Hyperscanning with functional near-infrared spectroscopy (fNIRS) is an emerging neuroimaging application that measures the nuanced neural signatures underlying social interactions.

no code implementations • 3 Nov 2021 • John Mern, Kyle Hatch, Ryan Silva, Cameron Hickert, Tamim Sookoor, Mykel J. Kochenderfer

The proposed deep reinforcement learning approach outperforms a fully automated playbook method in simulation, taking less disruptive actions while also defending more nodes on the network.

1 code implementation • NeurIPS 2021 • Phil Chen, Masha Itkina, Ransalu Senanayake, Mykel J. Kochenderfer

We evaluate our method on a variety of generative models, including variational autoencoders and auto-regressive architectures.

no code implementations • 16 Sep 2021 • John Mern, Sidhart Krishnan, Anil Yildiz, Kyle Hatch, Mykel J. Kochenderfer

In this work, we propose a method to build predictable policy trees as surrogates for policies such as neural networks.

1 code implementation • 5 Sep 2021 • Masha Itkina, Ye-Ji Mun, Katherine Driggs-Campbell, Mykel J. Kochenderfer

We propose an occlusion inference method that characterizes observed behaviors of human agents as sensor measurements, and fuses them with those from a standard sensor suite.

2 code implementations • 3 Aug 2021 • Chelsea Sidrane, Amir Maleki, Ahmed Irfan, Mykel J. Kochenderfer

In response to this challenge, we present OVERT: a sound algorithm for safety verification of nonlinear discrete-time closed loop dynamical systems with neural network control policies.

1 code implementation • 27 Jul 2021 • Mark Koren, Ahmed Nassar, Mykel J. Kochenderfer

Validating the safety of autonomous systems generally requires the use of high-fidelity simulators that adequately capture the variability of real-world scenarios.

1 code implementation • 23 Jul 2021 • Ransalu Senanayake, Kyle Beltran Hatch, Jason Zheng, Mykel J. Kochenderfer

This paper explores a Bayesian approach that captures our uncertainty in the map given training data.

no code implementations • 9 Jun 2021 • John Mern, Kyle Hatch, Ryan Silva, Jeff Brush, Mykel J. Kochenderfer

Defending computer networks from cyber attack requires coordinating actions across multiple nodes based on imperfect indicators of compromise while minimizing disruptions to network operations.

no code implementations • 9 Jun 2021 • Christopher A. Strong, Sydney M. Katz, Anthony L. Corso, Mykel J. Kochenderfer

We demonstrate how to formulate and solve three types of optimization problems: (i) minimization of any convex function over the output space, (ii) minimization of a convex function over the output of two networks in series with an adversarial perturbation in the layer between them, and (iii) maximization of the difference in output between two networks.

no code implementations • 8 Jun 2021 • John Mern, Mykel J. Kochenderfer

Monte Carlo planners can often return sub-optimal actions, even if they are guaranteed to converge in the limit of infinite samples.

no code implementations • 14 May 2021 • Sydney M. Katz, Anthony L. Corso, Christopher A. Strong, Mykel J. Kochenderfer

For this reason, recent work has focused on combining techniques in formal methods and reachability analysis to obtain guarantees on the closed-loop performance of neural network controllers.

1 code implementation • 5 May 2021 • Kunal Menda, Jayesh K. Gupta, Zachary Manchester, Mykel J. Kochenderfer

Structured Mechanical Models (SMMs) are a data-efficient black-box parameterization of mechanical systems, typically fit to data by minimizing the error between predicted and observed accelerations or next states.

1 code implementation • 1 Mar 2021 • Sydney M. Katz, Kyle D. Julian, Christopher A. Strong, Mykel J. Kochenderfer

In this work, we develop a method to use the results from neural network verification tools to provide probabilistic safety guarantees on a neural network controller.

no code implementations • 24 Feb 2021 • Sheng Li, Yutai Zhou, Ross Allen, Mykel J. Kochenderfer

Communication is a important factor that enables agents work cooperatively in multi-agent reinforcement learning (MARL).

Multi-agent Reinforcement Learning
reinforcement-learning
**+1**

1 code implementation • 12 Jan 2021 • Shushman Choudhury, Jayesh K. Gupta, Peter Morales, Mykel J. Kochenderfer

We also introduce a multi-drone delivery domain with dynamic, i. e., state-dependent coordination graphs, and demonstrate how our approach scales to large problems on this domain that are intractable for other MCTS methods.

no code implementations • 9 Dec 2020 • Anthony Corso, Mykel J. Kochenderfer

Safety validation is important during the development of safety-critical autonomous systems but can require significant computational effort.

no code implementations • 9 Nov 2020 • Xiaobai Ma, Jiachen Li, Mykel J. Kochenderfer, David Isele, Kikuo Fujimura

Deep reinforcement learning (DRL) provides a promising way for learning navigation in complex autonomous driving scenarios.

no code implementations • 4 Nov 2020 • Robert J. Moss, Ritchie Lee, Nicholas Visser, Joachim Hochwarth, James G. Lopez, Mykel J. Kochenderfer

To find failure events and their likelihoods in flight-critical systems, we investigate the use of an advanced black-box stress testing approach called adaptive stress testing.

no code implementations • 3 Nov 2020 • Julia Nitsch, Masha Itkina, Ransalu Senanayake, Juan Nieto, Max Schmidt, Roland Siegwart, Mykel J. Kochenderfer, Cesar Cadena

A mechanism to detect OOD samples is important for safety-critical applications, such as automotive perception, to trigger a safe fallback mode.

no code implementations • 20 Oct 2020 • Christopher Lazarus, James G. Lopez, Mykel J. Kochenderfer

The airworthiness and safety of a non-pedigreed autopilot must be verified, but the cost to formally do so can be prohibitive.

1 code implementation • 19 Oct 2020 • Bernard Lange, Masha Itkina, Mykel J. Kochenderfer

Safe and proactive planning in robotic systems generally requires accurate predictions of the environment.

1 code implementation • NeurIPS 2020 • Masha Itkina, Boris Ivanovic, Ransalu Senanayake, Mykel J. Kochenderfer, Marco Pavone

Discrete latent spaces in variational autoencoders have been shown to effectively capture the data distribution for many real-world problems such as natural language understanding, human intent prediction, and visual scene representation.

1 code implementation • 7 Oct 2020 • John Mern, Anil Yildiz, Larry Bush, Tapan Mukerji, Mykel J. Kochenderfer

Online solvers for partially observable Markov decision processes have difficulty scaling to problems with large action spaces.

no code implementations • 7 Oct 2020 • Christopher A. Strong, Haoze Wu, Aleksandar Zeljić, Kyle D. Julian, Guy Katz, Clark Barrett, Mykel J. Kochenderfer

However, individual "yes or no" questions cannot answer qualitative questions such as "what is the largest error within these bounds"; the answers to these lie in the domain of optimization.

1 code implementation • 7 Oct 2020 • John Mern, Anil Yildiz, Zachary Sunberg, Tapan Mukerji, Mykel J. Kochenderfer

Monte Carlo tree search with progressive widening attempts to improve scaling by sampling from the action space to construct a policy search tree.

no code implementations • NeurIPS 2020 • Andrea Zanette, Alessandro Lazaric, Mykel J. Kochenderfer, Emma Brunskill

There has been growing progress on theoretical analyses for provably efficient learning in MDPs with linear function approximation, but much of the existing work has made strong assumptions to enable exploration by conventional exploration frameworks.

no code implementations • 15 Aug 2020 • Duncan Eddy, Mykel J. Kochenderfer

This paper introduces a new approach for solving the satellite scheduling problem by generating an infeasibility-based graph representation of the problem and finding a maximal independent set of vertices for the graph.

no code implementations • 1 Jul 2020 • Ransalu Senanayake, Maneekwan Toyungyernsub, Mingyu Wang, Mykel J. Kochenderfer, Mac Schwager

We can use driving data collected over a long period of time to extract rich information about how vehicles behave in different areas of the roads.

1 code implementation • ICML 2020 • Kunal Menda, Jean de Becdelièvre, Jayesh K. Gupta, Ilan Kroo, Mykel J. Kochenderfer, Zachary Manchester

System identification is a key step for model-based control, estimator design, and output prediction.

1 code implementation • 19 Jun 2020 • Sheng Li, Jayesh K. Gupta, Peter Morales, Ross Allen, Mykel J. Kochenderfer

Coordination graph based formalization allows reasoning about the joint action based on the structure of interactions.

no code implementations • 17 Jun 2020 • John Mern, Peter Morales, Mykel J. Kochenderfer

The proposed method defines a conditional distribution for each variable in a sequential process by conditioning the parameters of a normalizing flow with recurrent neural connections.

1 code implementation • 16 Jun 2020 • Kyle Brown, Oriana Peltzer, Martin A. Sehr, Mac Schwager, Mykel J. Kochenderfer

We study the problem of sequential task assignment and collision-free routing for large teams of robots in applications with inter-task precedence constraints (e. g., task $A$ and task $B$ must both be completed before task $C$ may begin).

no code implementations • 15 Jun 2020 • Kyle Brown, Katherine Driggs-Campbell, Mykel J. Kochenderfer

We present a review and taxonomy of 200 models from the literature on driver behavior modeling.

1 code implementation • 27 May 2020 • Shushman Choudhury, Jayesh K. Gupta, Mykel J. Kochenderfer, Dorsa Sadigh, Jeannette Bohg

We consider the problem of dynamically allocating tasks to multiple agents under time window constraints and task completion uncertainty.

no code implementations • 25 May 2020 • Maxime Bouton, Alireza Nakhaei, David Isele, Kikuo Fujimura, Mykel J. Kochenderfer

This approach exposes the agent to a broad variety of behaviors during training, which promotes learning policies that are robust to model discrepancies.

no code implementations • 6 May 2020 • Anthony Corso, Robert J. Moss, Mark Koren, Ritchie Lee, Mykel J. Kochenderfer

Autonomous cyber-physical systems (CPS) can improve safety and efficiency for safety-critical applications, but require rigorous testing before deployment.

1 code implementation • 6 May 2020 • Erdem Biyik, Nicolas Huynh, Mykel J. Kochenderfer, Dorsa Sadigh

Our results in simulations and a user study suggest that our approach can efficiently learn expressive reward functions for robotics tasks.

1 code implementation • L4DC 2020 • Jayesh K. Gupta, Kunal Menda, Zachary Manchester, Mykel J. Kochenderfer

Deep neural networks have been used to learn models of robot dynamics from data, but they suffer from data-inefficiency and the difficulty to incorporate prior knowledge.

2 code implementations • 14 Apr 2020 • Anthony Corso, Mykel J. Kochenderfer

Our methodology is demonstrated for the safety validation of an autonomous vehicle in the context of an unprotected left turn and a crosswalk with a pedestrian.

no code implementations • 14 Apr 2020 • Anthony Corso, Ritchie Lee, Mykel J. Kochenderfer

In this work, we present a new safety validation approach that attempts to estimate the distribution over failures of an autonomous policy using approximate dynamic programming.

no code implementations • 8 Apr 2020 • Mark Koren, Mykel J. Kochenderfer

We demonstrate that GE is able to find failures without domain-specific heuristics, such as the distance between the car and the pedestrian, on scenarios that other RL techniques are unable to solve.

no code implementations • 8 Apr 2020 • Mark Koren, Anthony Corso, Mykel J. Kochenderfer

Validation is a key challenge in the search for safe autonomy.

no code implementations • 21 Mar 2020 • Shushman Choudhury, Nate Gruver, Mykel J. Kochenderfer

AIPPMS requires reasoning jointly about the effects of sensing and movement in terms of both energy expended and information gained.

no code implementations • 19 Mar 2020 • John Mern, Dorsa Sadigh, Mykel J. Kochenderfer

We show that our proposed representation results in an input space that is a factor of $m!$ smaller for inputs of $m$ objects.

1 code implementation • 11 Jan 2020 • Maxime Bouton, Jana Tumova, Mykel J. Kochenderfer

Autonomous systems are often required to operate in partially observable environments.

no code implementations • 23 Dec 2019 • Xiaobai Ma, Katherine Driggs-Campbell, Zongzhang Zhang, Mykel J. Kochenderfer

Gradient-based methods are often used for policy optimization in deep reinforcement learning, despite being vulnerable to local optima and saddle points.

1 code implementation • 15 Dec 2019 • Kyle D. Julian, Mykel J. Kochenderfer

The neural network outputs are bounded using neural network verification tools like Reluplex and Reluval, and a reachability method determines all possible ways aircraft encounters will resolve using neural network advisories and assuming bounded aircraft dynamics.

no code implementations • 14 Dec 2019 • Jeremy Morton, Freddie D. Witherden, Mykel J. Kochenderfer

The computational cost associated with simulating fluid flows can make it infeasible to run many simulations across multiple flow conditions.

no code implementations • NeurIPS 2019 • Andrea Zanette, Alessandro Lazaric, Mykel J. Kochenderfer, Emma Brunskill

We prove that if the features at any state can be represented as a convex combination of features at the anchor points, then errors are propagated linearly over iterations (instead of exponentially) and our method achieves a polynomial sample complexity bound in the horizon and the number of anchor points.

no code implementations • NeurIPS 2019 • Andrea Zanette, Mykel J. Kochenderfer, Emma Brunskill

This paper focuses on the problem of computing an $\epsilon$-optimal policy in a discounted Markov Decision Process (MDP) provided that we can access the reward and transition function through a generative model.

2 code implementations • 26 Sep 2019 • Shushman Choudhury, Kiril Solovey, Mykel J. Kochenderfer, Marco Pavone

Our results show that the framework computes solutions typically within a few seconds on commodity hardware, and that drones travel up to $360 \%$ of their flight range with public transit.

no code implementations • 25 Sep 2019 • Kunal Menda, Jean de Becdelièvre, Jayesh K Gupta, Ilan Kroo, Mykel J. Kochenderfer, Zachary Manchester

System identification is the process of building a mathematical model of an unknown system from measurements of its inputs and outputs.

1 code implementation • 2 Aug 2019 • Ross E. Allen, Jayesh K. Gupta, Jaime Pena, Yutai Zhou, Javona White Bear, Mykel J. Kochenderfer

This paper proposes a definition of system health in the context of multiple agents optimizing a joint reward function.

Multi-agent Reinforcement Learning
Policy Gradient Methods
**+2**

no code implementations • 2 Aug 2019 • Anthony Corso, Peter Du, Katherine Driggs-Campbell, Mykel J. Kochenderfer

Determining possible failure scenarios is a critical step in the evaluation of autonomous vehicle systems.

1 code implementation • 12 Jul 2019 • Sydney M. Katz, Anne-Claire Le Bihan, Mykel J. Kochenderfer

Airspace models have played an important role in the development and evaluation of aircraft collision avoidance systems for both manned and unmanned aircraft.

1 code implementation • 26 Jun 2019 • Maxime Bouton, Alireza Nakhaei, Kikuo Fujimura, Mykel J. Kochenderfer

In this work, we present a reinforcement learning approach to learn how to interact with drivers with different cooperation levels.

2 code implementations • 21 Jun 2019 • Shushman Choudhury, Mykel J. Kochenderfer

Sequential decision problems in applications such as manipulation in warehouses, multi-step meal preparation, and routing in autonomous vehicle networks often involve reasoning about uncertainty, planning over discrete modes as well as continuous states, and reacting to dynamic updates.

no code implementations • 6 May 2019 • Carl-Johan Hoel, Katherine Driggs-Campbell, Krister Wolff, Leo Laine, Mykel J. Kochenderfer

This paper introduces a general framework for tactical decision making, which combines the concepts of planning and learning, in the form of Monte Carlo tree search and deep reinforcement learning.

1 code implementation • 28 Apr 2019 • Masha Itkina, Katherine Driggs-Campbell, Mykel J. Kochenderfer

A key challenge for autonomous driving is safe trajectory planning in cluttered, urban environments with dynamic obstacles, such as pedestrians, bicyclists, and other vehicles.

1 code implementation • 25 Apr 2019 • Markus Schratter, Maxime Bouton, Mykel J. Kochenderfer, Daniel Watzenig

We show that combining the two approaches provides a robust autonomous braking system that reduces unnecessary braking caused by using the AEB system on its own.

2 code implementations • 25 Apr 2019 • Maxime Bouton, Alireza Nakhaei, Kikuo Fujimura, Mykel J. Kochenderfer

Navigating urban environments represents a complex task for automated vehicles.

no code implementations • 15 Apr 2019 • Maxime Bouton, Jesper Karlsson, Alireza Nakhaei, Kikuo Fujimura, Mykel J. Kochenderfer, Jana Tumova

We propose a generic approach to enforce probabilistic guarantees on an RL agent.

2 code implementations • 15 Mar 2019 • Changliu Liu, Tomer Arnon, Christopher Lazarus, Clark Barrett, Mykel J. Kochenderfer

Deep neural networks are widely used for nonlinear function approximation with applications ranging from computer vision to control.

1 code implementation • 14 Mar 2019 • Raunak P. Bhattacharyya, Derek J. Phillips, Changliu Liu, Jayesh K. Gupta, Katherine Driggs-Campbell, Mykel J. Kochenderfer

Recent developments in multi-agent imitation learning have shown promising results for modeling the behavior of human drivers.

no code implementations • 10 Mar 2019 • Edward Balaban, Stephen B. Johnson, Mykel J. Kochenderfer

Health management of complex dynamic systems has traditionally evolved separately from automated control, planning, and scheduling (generally referred to in the paper as decision making).

no code implementations • 8 Mar 2019 • Xiaobai Ma, Katherine Driggs-Campbell, Mykel J. Kochenderfer

To improve efficiency and reduce failures in autonomous vehicles, research has focused on developing robust and safe learning methods that take into account disturbances in the environment.

1 code implementation • 4 Mar 2019 • Bohan Wu, Jayesh K. Gupta, Mykel J. Kochenderfer

Learning interpretable and transferable subpolicies and performing task decomposition from a single, complex task is difficult.

1 code implementation • 26 Feb 2019 • Jeremy Morton, Freddie D. Witherden, Mykel J. Kochenderfer

We introduce the Deep Variational Koopman (DVK) model, a method for inferring distributions over observations that can be propagated linearly in time.

1 code implementation • 22 Feb 2019 • Jayesh K. Gupta, Kunal Menda, Zachary Manchester, Mykel J. Kochenderfer

We address the need for a flexible, gray-box model of mechanical systems that can seamlessly incorporate prior knowledge where it is available, and train expressive function approximators where it is not.

Model-based Reinforcement Learning
Reinforcement Learning (RL)
**+1**

1 code implementation • 5 Feb 2019 • Shushman Choudhury, Jacob P. Knickerbocker, Mykel J. Kochenderfer

We introduce the problem of Dynamic Real-time Multimodal Routing (DREAMR), which requires planning and executing routes under uncertainty for an autonomous agent.

no code implementations • 5 Feb 2019 • Mark Koren, Saud Alsaif, Ritchie Lee, Mykel J. Kochenderfer

This paper presents Monte Carlo Tree Search (MCTS) and Deep Reinforcement Learning (DRL) solutions that can scale to large environments.

no code implementations • 4 Feb 2019 • Derek J. Phillips, Juan Carlos Aragon, Anjali Roychowdhury, Regina Madigan, Sunil Chintakindi, Mykel J. Kochenderfer

Many automotive applications, such as Advanced Driver Assistance Systems (ADAS) for collision avoidance and warnings, require estimating the future automotive risk of a driving scene.

no code implementations • 10 Dec 2018 • John Mern, Kyle Julian, Rachael E. Tompa, Mykel J. Kochenderfer

A reliable sense-and-avoid system is critical to enabling safe autonomous operation of unmanned aircraft.

no code implementations • 25 Nov 2018 • Andrea Zanette, Junzi Zhang, Mykel J. Kochenderfer

This paper focuses on the problem of determining as large a region as possible where a function exceeds a given threshold with high probability.

no code implementations • 6 Nov 2018 • Ritchie Lee, Ole J. Mengshoel, Anshu Saksena, Ryan Gardner, Daniel Genin, Joshua Silbermann, Michael Owen, Mykel J. Kochenderfer

Finding the most likely path to a set of failure states is important to the analysis of safety-critical systems that operate over a sequence of time steps, such as aircraft collision avoidance systems and autonomous cars.

no code implementations • 9 Oct 2018 • Kyle D. Julian, Mykel J. Kochenderfer

Teams of autonomous unmanned aircraft can be used to monitor wildfires, enabling firefighters to make informed decisions.

no code implementations • 9 Oct 2018 • Kyle D. Julian, Mykel J. Kochenderfer, Michael P. Owen

One approach to designing decision making logic for an aircraft collision avoidance system frames the problem as a Markov decision process and optimizes the system using dynamic programming.

1 code implementation • 4 Oct 2018 • Kyle D. Julian, Mykel J. Kochenderfer

The second approach uses a particle filter to predict wildfire growth and uses observations to estimate uncertainties relating to wildfire expansion.

no code implementations • 26 Sep 2018 • Louis Dressel, Mykel J. Kochenderfer

A common method uses information maps that estimate the value of taking measurements from any point in the sensor state space.

no code implementations • 22 Jul 2018 • Kunal Menda, Katherine Driggs-Campbell, Mykel J. Kochenderfer

While imitation learning is often used in robotics, the approach frequently suffers from data mismatch and compounding errors.

no code implementations • NeurIPS 2018 • Rui Shu, Hung H. Bui, Shengjia Zhao, Mykel J. Kochenderfer, Stefano Ermon

In this paper, we leverage the fact that VAEs rely on amortized inference and propose techniques for amortized inference regularization (AIR) that control the smoothness of the inference model.

1 code implementation • NeurIPS 2018 • Jeremy Morton, Freddie D. Witherden, Antony Jameson, Mykel J. Kochenderfer

The design of flow control systems remains a challenge due to the nonlinear nature of the equations that govern fluid flow.

1 code implementation • 2 Mar 2018 • Raunak P. Bhattacharyya, Derek J. Phillips, Blake Wulfe, Jeremy Morton, Alex Kuefler, Mykel J. Kochenderfer

Simulation is an appealing option for validating the safety of autonomous vehicles.

1 code implementation • 6 Feb 2018 • Maxime Bouton, Kyle Julian, Alireza Nakhaei, Kikuo Fujimura, Mykel J. Kochenderfer

In contexts where an agent interacts with multiple entities, utility decomposition can be used to separate the global objective into local tasks considering each individual entity independently.

no code implementations • 5 Feb 2018 • Blake Wulfe, Sunil Chintakindi, Sou-Cheng T. Choi, Rory Hartong-Redden, Anuradha Kodali, Mykel J. Kochenderfer

Advanced collision avoidance and driver hand-off systems can benefit from the ability to accurately predict, in real time, the probability a vehicle will be involved in a collision within an intermediate horizon of 10 to 20 seconds.

no code implementations • 13 Oct 2017 • Alex Kuefler, Mykel J. Kochenderfer

Recent work on imitation learning has generated policies that reproduce expert behavior from multi-modal data.

1 code implementation • 19 Sep 2017 • Kunal Menda, Yi-Chun Chen, Justin Grana, James W. Bono, Brendan D. Tracey, Mykel J. Kochenderfer, David Wolpert

The incorporation of macro-actions (temporally extended actions) into multi-agent decision problems has the potential to address the curse of dimensionality associated with such decision problems.

no code implementations • 18 Sep 2017 • Kunal Menda, Katherine Driggs-Campbell, Mykel J. Kochenderfer

While imitation learning is becoming common practice in robotics, this approach often suffers from data mismatch and compounding errors.

no code implementations • 8 Sep 2017 • Guy Katz, Clark Barrett, David L. Dill, Kyle Julian, Mykel J. Kochenderfer

Autonomous vehicles are highly complex systems, required to function reliably in a wide variety of situations.

no code implementations • 30 Aug 2017 • Ritchie Lee, Mykel J. Kochenderfer, Ole J. Mengshoel, Joshua Silbermann

In particular, when a grammar based on temporal logic is used, we show that GBDTs can be used for the interpretable classi cation of high-dimensional and heterogeneous time series data.

no code implementations • 25 Jul 2017 • Jeremy Morton, Tim A. Wheeler, Mykel J. Kochenderfer

Manufacturers of safety-critical systems must make the case that their product is sufficiently safe for public deployment.

2 code implementations • 19 Apr 2017 • Jeremy Morton, Mykel J. Kochenderfer

In this work, we propose a method for learning driver models that account for variables that cannot be observed directly.

no code implementations • 14 Apr 2017 • Maxime Bouton, Akansel Cosgun, Mykel J. Kochenderfer

Urban intersections represent a complex environment for autonomous vehicles with many sources of uncertainty.

Robotics

no code implementations • 29 Nov 2015 • Philipp Robbel, Frans A. Oliehoek, Mykel J. Kochenderfer

We present an approach to mitigate this limitation for certain types of multiagent systems, exploiting a property that can be thought of as "anonymous influence" in the factored MDP.

no code implementations • 9 Aug 2014 • Erik J. Schlicht, Ritchie Lee, David H. Wolpert, Mykel J. Kochenderfer, Brendan Tracey

Multi-fidelity methods combine inexpensive low-fidelity simulations with costly but highfidelity simulations to produce an accurate model of a system of interest at minimal cost.

no code implementations • 21 May 2014 • Dimitris Bertsimas, J. Daniel Griffith, Vishal Gupta, Mykel J. Kochenderfer, Velibor V. Mišić, Robert Moss

In this paper, we adapt both MCTS and MO to a problem inspired by tactical wildfire and management and undertake an extensive computational study comparing the two methods on large scale instances in terms of both the state and the action spaces.

Cannot find the paper you are looking for? You can
Submit a new open access paper.

Contact us on:
hello@paperswithcode.com
.
Papers With Code is a free resource with all data licensed under CC-BY-SA.