Search Results for author:
Found 42 papers, 16 papers with code
PaRCE: Probabilistic and Reconstruction-based Competency Estimation for CNN-based Image Classification
Convolutional neural networks (CNNs) are extremely popular and effective for image classification tasks but tend to be overly confident in their predictions.
What Do Learning Dynamics Reveal About Generalization in LLM Reasoning?
We find that a model's generalization behavior can be effectively characterized by a training metric we call pre-memorization train accuracy: the accuracy of model samples on training queries before they begin to copy the exact reasoning steps from the training set.
Competency-Aware Planning for Probabilistically Safe Navigation Under Perception Uncertainty
In an effort to enable safe navigation under perception uncertainty, we develop a probabilistic and reconstruction-based competency estimation (PaRCE) method to estimate the model's level of familiarity with an input image as a whole and with specific regions in the image.
Certifiable Reachability Learning Using a New Lipschitz Continuous Value Function
Our framework has two main parts: offline learning of a newly designed reachavoid value function, and post-learning certification.
Effects of dynamic power electronic load models on power systems analysis using ZIP-E loads
These results suggests that the constant power portion of the ZIP load has a large destabilizing effect and can generally overestimate instability, and that attention should be drawn to load model choice if operating near a stability boundary.
Understanding the Dependence of Perception Model Competency on Regions in an Image
We find that the competency gradients and reconstruction loss methods show great promise in identifying regions associated with low model competency, particularly when aspects of the image that are unfamiliar to the perception model are causing this reduction in competency.
Stranger Danger! Identifying and Avoiding Unpredictable Pedestrians in RL-based Social Robot Navigation
Reinforcement learning (RL) methods for social robot navigation show great success navigating robots through large crowds of people, but the performance of these learning-based methods tends to degrade in particularly challenging or unfamiliar situations due to the models' dependency on representative training data.
Unfamiliar Finetuning Examples Control How Language Models Hallucinate
Leveraging our previous observations on controlling hallucinations, we propose an approach for learning more reliable reward models, and show that they improve the efficacy of RL factuality finetuning in long-form biography and book/movie plot generation tasks.
Intent Demonstration in General-Sum Dynamic Games via Iterative Linear-Quadratic Approximations
Autonomous agents should be able to coordinate with other agents without knowing their intents ahead of time.
Hacking Predictors Means Hacking Cars: Using Sensitivity Analysis to Identify Trajectory Prediction Vulnerabilities for Autonomous Driving Security
Adversarial attacks on learning-based multi-modal trajectory predictors have already been demonstrated.
Deep Neural Networks Tend To Extrapolate Predictably
Rather than extrapolating in arbitrary ways, we observe that neural network predictions often tend towards a constant value as input data becomes increasingly OOD.
Linking vision and motion for self-supervised object-centric perception
Object-centric representations enable autonomous driving algorithms to reason about interactions between many independent agents and scene features.
Scenario-Game ADMM: A Parallelized Scenario-Based Solver for Stochastic Noncooperative Games
To accommodate this, we decompose the approximated game into a set of smaller games with few constraints for each sampled scenario, and propose a decentralized, consensus-based ADMM algorithm to efficiently compute a generalized Nash equilibrium (GNE) of the approximated game.
Multi-Agent Reachability Calibration with Conformal Prediction
We investigate methods to provide safety assurances for autonomous agents that incorporate predictions of other, uncontrolled agents' behavior into their own trajectory planning.
Multi-Task Imitation Learning for Linear Dynamical Systems
In particular, we consider a setting where learning is split into two phases: (a) a pre-training step where a shared $k$-dimensional representation is learned from $H$ source policies, and (b) a target policy fine-tuning step where the learned representation is used to parameterize the policy class.
Lyapunov Density Models: Constraining Distribution Shift in Learning-Based Control
Learned models and policies can generalize effectively when evaluated within the distribution of the training data, but can produce unpredictable and erroneous outputs on out-of-distribution inputs.
Navigation between initial and desired community states using shortcuts
Using several empirical datasets, we show that (1) non-brute-force navigation is only possible between some state pairs, (2) shortcuts exist between many state pairs; and (3) changes in abundance and richness are the strongest predictors of shortcut existence, independent of dataset and algorithm choices.
Inducing Structure in Reward Learning by Learning Features
To get around this issue, recent deep Inverse Reinforcement Learning (IRL) methods learn rewards directly from the raw state but this is challenging because the robot has to implicitly learn the features that are important and how to combine them, simultaneously.
Learning from learning machines: a new generation of AI technology to meet the needs of science
We outline emerging opportunities and challenges to enhance the utility of AI for scientific discovery.
Incorporating Data Uncertainty in Object Tracking Algorithms
However, for detections generated from neural-network processed camera inputs, these measurement error statistics are not sufficient to represent the primary source of errors, namely a dissimilarity between run-time sensor input and the training data upon which the detector was trained.
Multi-Task Learning with Sequence-Conditioned Transporter Networks
Enabling robots to solve multiple manipulation tasks has a wide range of industrial applications.
Testing for Typicality with Respect to an Ensemble of Learned Distributions
In particular, we propose training an ensemble of density models, considering data to be anomalous if the data is anomalous with respect to any member of the ensemble.
DeepReach: A Deep Learning Approach to High-Dimensional Reachability
Its advantages include compatibility with general nonlinear system dynamics, formal treatment of bounded disturbances, and the ability to deal with state and input constraints.
Expert Selection in High-Dimensional Markov Decision Processes
In this work we present a multi-armed bandit framework for online expert selection in Markov decision processes and demonstrate its use in high-dimensional settings.
Feature Expansive Reward Learning: Rethinking Human Input
When the correction cannot be explained by these features, recent work in deep Inverse Reinforcement Learning (IRL) suggests that the robot could ask for task demonstrations and recover a reward defined over the raw state space.
Visual Navigation Among Humans with Optimal Control as a Supervisor
Videos describing our approach and experiments, as well as a demo of HumANav are available on the project website.
Generating Robust Supervision for Learning-Based Visual Navigation Using Hamilton-Jacobi Reachability
In Bansal et al. (2019), a novel visual navigation framework that combines learning-based and model-based approaches has been proposed.
Combining Optimal Control and Learning for Visual Navigation in Novel Environments
Model-based control is a popular paradigm for robot navigation because it can leverage a known dynamics model to efficiently plan robust robot trajectories.
Regression-based Inverter Control for Decentralized Optimal Power Flow and Voltage Regulation
Electronic power inverters are capable of quickly delivering reactive power to maintain customer voltages within operating tolerances and to reduce system losses in distribution grids.
Fast Neural Network Verification via Shadow Prices
To use neural networks in safety-critical settings it is paramount to provide assurances on their runtime operation.
A Successive-Elimination Approach to Adaptive Robotic Sensing
We study an adaptive source seeking problem, in which a mobile robot must identify the strongest emitter(s) of a signal in an environment with background emissions.
Towards Distributed Energy Services: Decentralizing Optimal Power Flow with Machine Learning
We consider distribution systems with multiple controllable Distributed Energy Resources (DERs) and present a data-driven approach to learn control policies for each DER to reconstruct and mimic the solution to a centralized OPF problem from solely locally available information.
Countering Feedback Delays in Multi-Agent Learning
We consider a model of game-theoretic learning based on online mirror descent (OMD) with asynchronous and delayed feedback information.
On Identification of Distribution Grids
Large-scale integration of distributed energy resources into residential distribution feeders necessitates careful control of their operation through power flow analysis.
A Sequential Approximation Framework for Coded Distributed Optimization
As an application of the results, we demonstrate solving optimization problems using a sequential approximation approach, which accelerates the algorithm in a distributed system with stragglers.
MBMF: Model-Based Priors for Model-Free Reinforcement Learning
Reinforcement Learning is divided in two main paradigms: model-free and model-based.
Fully Decentralized Policies for Multi-Agent Systems: An Information Theoretic Approach
Learning cooperative policies for multi-agent systems is often challenged by partial observability and a lack of coordination.
A Multi-Armed Bandit Approach for Online Expert Selection in Markov Decision Processes
We formulate a multi-armed bandit (MAB) approach to choosing expert policies online in Markov decision processes (MDPs).
Systems and Control
Minimizing Regret on Reflexive Banach Spaces and Nash Equilibria in Continuous Zero-Sum Games
We study a general adversarial online learning problem, in which we are given a decision set X' in a reflexive Banach space X and a sequence of reward vectors in the dual space of X.
Recursive Regression with Neural Networks: Approximating the HJI PDE Solution
The majority of methods used to compute approximations to the Hamilton-Jacobi-Isaacs partial differential equation (HJI PDE) rely on the discretization of the state space to perform dynamic programming updates.
Inverse Power Flow Problem
We show that the admittance matrix can be uniquely identified from a sequence of measurements corresponding to different steady states when every node in the system is equipped with a measurement device, and a Kron-reduced admittance matrix can be determined even if some nodes in the system are not monitored (hidden nodes).
Minimizing Regret on Reflexive Banach Spaces and Learning Nash Equilibria in Continuous Zero-Sum Games
Under the assumption of uniformly continuous rewards, we obtain explicit anytime regret bounds in a setting where the decision set is the set of probability distributions on a compact metric space $S$ whose Radon-Nikodym derivatives are elements of $L^p(S)$ for some $p > 1$.
