no code implementations • 12 Dec 2024 • Paul Brunzema, Mikkel Jordahn, John Willes, Sebastian Trimpe, Jasper Snoek, James Harrison
While Bayesian neural networks (BNNs) are a promising direction for higher capacity surrogate models, they have so far seen limited use due to poor performance on some problem types.
no code implementations • 22 Nov 2024 • Teodor Alexandru Szente, James Harrison, Mihai Zanfir, Cristian Sminchisescu
Fractional gradient descent has been studied extensively, with a focus on its ability to extend traditional gradient descent methods by incorporating fractional-order derivatives.
no code implementations • 10 Oct 2024 • Carolin Schmidt, Daniele Gammelli, James Harrison, Marco Pavone, Filipe Rodrigues
In this work, we propose OHIO: a framework for offline reinforcement learning (RL) of hierarchical policies.
2 code implementations • 17 Apr 2024 • James Harrison, John Willes, Jasper Snoek
We introduce a deterministic variational formulation for training Bayesian last layer neural networks.
1 code implementation • 15 Feb 2024 • Tobias Enders, James Harrison, Maximilian Schiffer
We study the robustness of deep reinforcement learning algorithms against distribution shifts within contextual multi-stage stochastic combinatorial optimization problems from the operations research domain.
1 code implementation • 7 Feb 2024 • Allan Zhou, Chelsea Finn, James Harrison
A challenging problem in many modern machine learning tasks is to process weight-space features, i. e., to transform or extract information from the weights and gradients of a neural network.
no code implementations • 11 Dec 2023 • Avi Singh, John D. Co-Reyes, Rishabh Agarwal, Ankesh Anand, Piyush Patil, Xavier Garcia, Peter J. Liu, James Harrison, Jaehoon Lee, Kelvin Xu, Aaron Parisi, Abhishek Kumar, Alex Alemi, Alex Rizkowsky, Azade Nova, Ben Adlam, Bernd Bohnet, Gamaleldin Elsayed, Hanie Sedghi, Igor Mordatch, Isabelle Simpson, Izzeddin Gur, Jasper Snoek, Jeffrey Pennington, Jiri Hron, Kathleen Kenealy, Kevin Swersky, Kshiteej Mahajan, Laura Culp, Lechao Xiao, Maxwell L. Bileschi, Noah Constant, Roman Novak, Rosanne Liu, Tris Warkentin, Yundi Qian, Yamini Bansal, Ethan Dyer, Behnam Neyshabur, Jascha Sohl-Dickstein, Noah Fiedel
To do so, we investigate a simple self-training method based on expectation-maximization, which we call ReST$^{EM}$, where we (1) generate samples from the model and filter them using binary feedback, (2) fine-tune the model on these samples, and (3) repeat this process a few times.
1 code implementation • 16 May 2023 • Daniele Gammelli, James Harrison, Kaidi Yang, Marco Pavone, Filipe Rodrigues, Francisco C. Pereira
Optimization problems over dynamic networks have been extensively studied and widely used in the past decades to formulate numerous real-world problems.
1 code implementation • NeurIPS 2023 • Oscar Li, James Harrison, Jascha Sohl-Dickstein, Virginia Smith, Luke Metz
Unrolled computation graphs are prevalent throughout machine learning but present challenges to automatic differentiation (AD) gradient estimation methods when their loss functions exhibit extreme local sensitivtiy, discontinuity, or blackbox characteristics.
1 code implementation • 14 Dec 2022 • Tobias Enders, James Harrison, Marco Pavone, Maximilian Schiffer
We consider the sequential decision-making problem of making proactive request assignment and rejection decisions for a profit-maximizing operator of an autonomous mobility on demand system.
no code implementations • 8 Dec 2022 • Louis Kirsch, James Harrison, Jascha Sohl-Dickstein, Luke Metz
We further show that the capabilities of meta-trained algorithms are bottlenecked by the accessible state size (memory) determining the next prediction, unlike standard models which are thought to be bottlenecked by parameter count.
no code implementations • 2 Dec 2022 • Rohan Sinha, James Harrison, Spencer M. Richards, Marco Pavone
We propose a learning-based robust predictive control algorithm that compensates for significant uncertainty in the dynamics for a class of discrete-time systems that are nominally linear with an additive nonlinear component.
1 code implementation • 17 Nov 2022 • Luke Metz, James Harrison, C. Daniel Freeman, Amil Merchant, Lucas Beyer, James Bradbury, Naman Agrawal, Ben Poole, Igor Mordatch, Adam Roberts, Jascha Sohl-Dickstein
While deep learning models have replaced hand-designed features across many domains, these models are still trained with hand-designed optimizers.
2 code implementations • 23 Sep 2022 • Boris Ivanovic, James Harrison, Marco Pavone
Learning-based behavior prediction methods are increasingly being deployed in real-world autonomous systems, e. g., in fleets of self-driving vehicles, which are beginning to commercially operate in major cities across the world.
1 code implementation • 22 Sep 2022 • James Harrison, Luke Metz, Jascha Sohl-Dickstein
We apply the resulting learned optimizer to a variety of neural network training tasks, where it outperforms the current state of the art learned optimizer -- at matched optimizer computational overhead -- with regard to optimization performance and meta-training speed, and is capable of generalization to tasks far different from those it was meta-trained on.
1 code implementation • 22 Mar 2022 • Luke Metz, C. Daniel Freeman, James Harrison, Niru Maheswaranathan, Jascha Sohl-Dickstein
We further leverage our analysis to construct a learned optimizer that is both faster and more memory efficient than previous work.
1 code implementation • 15 Feb 2022 • Daniele Gammelli, Kaidi Yang, James Harrison, Filipe Rodrigues, Francisco C. Pereira, Marco Pavone
Autonomous Mobility-on-Demand (AMoD) systems represent an attractive alternative to existing transportation paradigms, currently challenged by urbanization and increasing travel needs.
no code implementations • 11 Nov 2021 • Thomas Lew, Apoorva Sharma, James Harrison, Edward Schmerling, Marco Pavone
We identify an issue in recent approaches to learning-based control that reformulate systems with uncertain dynamics using a stochastic differential equation.
no code implementations • 29 Jul 2021 • John Willes, James Harrison, Ali Harakeh, Chelsea Finn, Marco Pavone, Steven Waslander
As autonomous decision-making agents move from narrow operating environments to unstructured worlds, learning systems must move from a closed-world formulation to an open-world and few-shot setting in which agents continuously learn new classes from small amounts of information.
1 code implementation • 23 Apr 2021 • Daniele Gammelli, Kaidi Yang, James Harrison, Filipe Rodrigues, Francisco C. Pereira, Marco Pavone
Autonomous mobility-on-demand (AMoD) systems represent a rapidly developing mode of transportation wherein travel requests are dynamically handled by a coordinated fleet of robotic, self-driving vehicles.
no code implementations • 16 Apr 2021 • Rohan Sinha, James Harrison, Spencer M. Richards, Marco Pavone
We propose a learning-based robust predictive control algorithm that compensates for significant uncertainty in the dynamics for a class of discrete-time systems that are nominally linear with an additive nonlinear component.
no code implementations • 6 Apr 2021 • Robert Dyro, James Harrison, Apoorva Sharma, Marco Pavone
As robotic systems move from highly structured environments to open worlds, incorporating uncertainty from dynamics learning or state estimation into the control pipeline is essential for robust performance.
no code implementations • 9 Nov 2020 • Jinghe Zhang, Kamran Kowsari, Mehdi Boukhechba, James Harrison, Jennifer Lobo, Laura Barnes
Chronic kidney disease (CKD) is a gradual loss of renal function over time, and it increases the risk of mortality, decreased quality of life, as well as serious complications.
no code implementations • 26 Aug 2020 • Thomas Lew, Apoorva Sharma, James Harrison, Andrew Bylard, Marco Pavone
In this work, we propose a practical and theoretically-justified approach to maintaining safety in the presence of dynamics uncertainty.
no code implementations • ICML Workshop LifelongML 2020 • Annie Xie, James Harrison, Chelsea Finn
As humans, our goals and our environment are persistently changing throughout our lifetime based on our experiences, actions, and internal and external drives.
2 code implementations • NeurIPS 2020 • James Harrison, Apoorva Sharma, Chelsea Finn, Marco Pavone
In this work, we enable the application of generic meta-learning algorithms to settings where this task segmentation is unavailable, such as continual online learning with a time-varying task.
no code implementations • 15 Feb 2019 • Sandeep Chinchali, Apoorva Sharma, James Harrison, Amine Elhafsi, Daniel Kang, Evgenya Pergament, Eyal Cidon, Sachin Katti, Marco Pavone
In this paper, we formulate a novel Robot Offloading Problem --- how and when should robots offload sensing tasks, especially if they are uncertain, to improve accuracy while minimizing the cost of cloud communication?
no code implementations • 9 Jan 2019 • Apoorva Sharma, James Harrison, Matthew Tsao, Marco Pavone
The first, RAMCP-F, converges to an optimal risk-sensitive policy without having to rebuild the search tree as the underlying belief over models is perturbed.
3 code implementations • 24 Jul 2018 • James Harrison, Apoorva Sharma, Marco Pavone
However, this approach suffers from two main drawbacks: (1) it is computationally inefficient, as computation scales poorly with the number of samples; and (2) it can be data inefficient, as encoding prior knowledge that can aid the model through the choice of kernel and associated hyperparameters is often challenging and unintuitive.
1 code implementation • 16 Jun 2018 • Boris Ivanovic, James Harrison, Apoorva Sharma, Mo Chen, Marco Pavone
Our Backward Reachability Curriculum (BaRC) begins policy training from states that require a small number of actions to accomplish the task, and expands the initial state distribution backwards in a dynamically-consistent manner once the policy optimization algorithm demonstrates sufficient performance.
2 code implementations • 16 Sep 2017 • Brian Ichter, James Harrison, Marco Pavone
This paper proposes a methodology for non-uniform sampling, whereby a sampling distribution is learned from demonstrations, and then used to bias sampling.