Task-Driven Estimation and Control via Information Bottlenecks

Pacelli, Vincent; Majumdar, Anirudha

doi:10.1109/icra.2019.8794213

Cited by 9 publications

(10 citation statements)

References 66 publications

(76 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In direct data-driven control design, we would like to use the input-output behavior data matrix " U J Y J ‰ J directly in place of the model p P in (7). The key idea here is to avoid identifying a model of input-output behaviors and to directly search for the optimal behavior for the task (2) within the span of observed behaviors contained in the data matrix " U J Y J ‰ J .…”

Section: A Direct Data-driven Controlmentioning

confidence: 99%

See 1 more Smart Citation

On Direct vs Indirect Data-Driven Predictive Control

Krishnan¹,

Pasqualetti²

2021

Preprint

View full text Add to dashboard Cite

In this work, we compare the direct and indirect approaches to data-driven predictive control of stochastic linear time-invariant systems. The distinction between the two approaches lies in the fact that the indirect approach involves identifying a lower dimensional model from data which is then used in a certainty-equivalent control design, while the direct approach avoids this intermediate step altogether. Working within an optimization-based framework, we find that the suboptimality gap measuring the control performance w.r.t. the optimal model-based control design vanishes with the size of the dataset only with the direct approach. The indirect approach has a higher rate of convergence, but its suboptimality gap does not vanish as the size of the dataset increases. This reveals the existence of two distinct regimes of performance as the size of the dataset of input-output behaviors is increased. We show that the indirect approach, by relying on the identification of a lower dimensional model, has lower variance and outperforms the direct approach for smaller datasets, while it incurs an asymptotic bias as a result of the process noise and a (possibly) incorrect assumption on the order of the identified model. The direct approach, however, does not incur an asymptotic bias, and outperforms the indirect approach for larger datasets. Ultimately, by revealing the existence of two non-asymptotic regimes for the performance of direct and indirect data-driven predictive control designs, our study suggests that neither approach is invariably superior and that the choice of design must, in practice, be informed by the available dataset.

show abstract

Section: A Direct Data-driven Controlmentioning

confidence: 99%

“…The above is the sense in which the direct data-driven control design formulation employs implicitly an estimate p P direct of the true behavior model P as stated earlier, and can be connected to the general data-driven control design formulation (7). We now obtain the minimizer p u direct in (10) as:…”

Section: A Direct Data-driven Controlmentioning

confidence: 99%

On Direct vs Indirect Data-Driven Predictive Control

Krishnan¹,

Pasqualetti²

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Literature Review: Recent work has applied information bottleneck theory [32] to build controllers that focus on actionable, task-relevant visual inputs for robust, generalizable navigation and grasping policies [27,26,29]. In contrast, we introduce a novel algorithm for co-designing communication and machine perception, which uses pre-trained task modules to learn salient, efficiently-computable representations.…”

Section: Train-time Onlymentioning

confidence: 99%

Co-Design of Communication and Machine Inference for Cloud Robotics

Nakanoya

Chinchali

Anemogiannis

et al. 2021

Robotics: Science and Systems XVII

View full text Add to dashboard Cite

Today, even the most compute-and-power constrained robots can measure complex, high data-rate video and LIDAR sensory streams. Often, such robots, ranging from lowpower drones to space and subterranean rovers, need to transmit high-bitrate sensory data to a remote compute server if they are uncertain or cannot scalably run complex perception or mapping tasks locally. However, today's representations for sensory data are mostly designed for human, not robotic, perception and thus often waste precious compute or wireless network resources to transmit unimportant parts of a scene that are unnecessary for a high-level robotic task. This paper presents an algorithm to learn task-relevant representations of sensory data that are codesigned with a pre-trained robotic perception model's ultimate objective. Our algorithm aggressively compresses robotic sensory data by up to 11 × more than competing methods. Further, it achieves high accuracy and robust generalization on diverse tasks including Mars terrain classification with low-power deep learning accelerators, neural motion planning, and environmental timeseries classification.

show abstract

“…A different line of work learns task-centric memory representations via information bottlenecks [22,23]. These approaches seek policies with "low complexity" as defined in terms of the information contained in the memory representation.…”

Section: Related Workmentioning

confidence: 99%

“…These approaches seek policies with "low complexity" as defined in terms of the information contained in the memory representation. For example, in [23], the objective is to minimize the information content about the state in the memory representation. Our work, instead, defines memory complexity in terms of the dimension; such a measure of complexity is more physically meaningful and tied to the robotic system's onboard memory constraints.…”

Section: Related Workmentioning

confidence: 99%

Learning to Actively Reduce Memory Requirements for Robot Control Tasks

Booker¹,

Majumdar²

2020

Preprint

Self Cite

View full text Add to dashboard Cite

Robots equipped with rich sensing modalities (e.g., RGB-D cameras) performing long-horizon tasks motivate the need for policies that are highly memory-efficient. State-of-the-art approaches for controlling robots often use memory representations that are excessively rich for the task or rely on handcrafted tricks for memory efficiency. Instead, this work provides a general approach for jointly synthesizing memory representations and policies; the resulting policies actively seek to reduce memory requirements (i.e., take actions that reduce memory usage). Specifically, we present a reinforcement learning framework that leverages an implementation of the group LASSO regularization to synthesize policies that employ low-dimensional and task-centric memory representations. We demonstrate the efficacy of our approach with simulated examples including navigation in discrete and continuous spaces as well as vision-based indoor navigation set in a photo-realistic simulator. The results on these examples indicate that our method is capable of finding policies that rely only on low-dimensional memory representations and actively reduce memory requirements.

show abstract

Task-Driven Estimation and Control via Information Bottlenecks

Cited by 9 publications

References 66 publications

On Direct vs Indirect Data-Driven Predictive Control

On Direct vs Indirect Data-Driven Predictive Control

Co-Design of Communication and Machine Inference for Cloud Robotics

Learning to Actively Reduce Memory Requirements for Robot Control Tasks

Contact Info

Product

Resources

About