Continuous-time neural networks are a class of machine learning systems that can tackle representation learning on spatiotemporal decision-making tasks. These models are typically represented by continuous differential equations. However, their expressive power when they are deployed on computers is bottlenecked by numerical differential equation solvers. This limitation has notably slowed down the scaling and understanding of numerous natural physical phenomena such as the dynamics of nervous systems. Ideally, we would circumvent this bottleneck by solving the given dynamical system in closed form. This is known to be intractable in general. Here, we show that it is possible to closely approximate the interaction between neurons and synapses—the building blocks of natural and artificial neural networks—constructed by liquid time-constant networks efficiently in closed form. To this end, we compute a tightly bounded approximation of the solution of an integral appearing in liquid time-constant dynamics that has had no known closed-form solution so far. This closed-form solution impacts the design of continuous-time and continuous-depth neural models. For instance, since time appears explicitly in closed form, the formulation relaxes the need for complex numerical solvers. Consequently, we obtain models that are between one and five orders of magnitude faster in training and inference compared with differential equation-based counterparts. More importantly, in contrast to ordinary differential equation-based continuous networks, closed-form networks can scale remarkably well compared with other deep learning instances. Lastly, as these models are derived from liquid networks, they show good performance in time-series modelling compared with advanced recurrent neural network models.
Abstract-The successful deployment of many autonomous systems in part hinges on providing rigorous guarantees on their performance and safety through a formal verification method, such as reachability analysis. In this work, we present a simple-to-implement, sampling-based algorithm for reachability analysis that is provably optimal up to any desired approximation accuracy. Our method achieves computational efficiency by judiciously sampling a finite subset of the state space and generating an approximate reachable set by conducting reachability analysis on this finite set of states. We prove that the reachable set generated by our algorithm approximates the ground-truth reachable set for any user-specified approximation accuracy. As a corollary to our main method, we introduce an asymptoticallyoptimal, anytime algorithm for reachability analysis. We present simulation results that reaffirm the theoretical properties of our algorithm and demonstrate its effectiveness in real-world inspired scenarios.I. INTRODUCTION Autonomous and highly automated systems inherently depend on effectively incorporating rigorous guarantees on the performance and safety through formal verification and validation methods. For instance, in order to ensure collision-free paths, advanced driver-assistance systems need to be capable of anticipating all potential actions of the driver without overly conservative assumptions. This requires performing on-line reachability analysis, i.e., computation of states that these vehicles can reach within a given time interval. It can also serve as a supervisory mechanism for any motion planner that incorporates deep learning. Going beyond the realm of autonomous driving, reachability analysis has shown promise as a tool for formal verification of a wide variety of systems. Applications of reachability analysis include safety, correctness, and controller synthesis problems involving intricate specifications or robotic systems such as autonomous aircraft and cars, medical robots, and personal-assistance robots.Typically the state of a system is not fully observable, e.g., a car might not have precise knowledge about its position. Thus, conducting accurate reachability analysis by definition requires reasoning about all possible trajectories from every possible state. Reasoning about all possible behaviors of a system renders reachability analysis computationally intractable in practice [3]. This computational challenge is further compounded by the generally large size and high complexity of the system in consideration, and the practical need to obtain verification results in a reasonably short time (i.e., seconds or minutes, not days) for the sake of, for example,
We present a provable, sampling-based approach for generating compact Convolutional Neural Networks (CNNs) by identifying and removing redundant filters from an over-parameterized network. Our algorithm uses a small batch of input data points to assign a saliency score to each filter and constructs an importance sampling distribution where filters that highly affect the output are sampled with correspondingly high probability. In contrast to existing filter pruning approaches, our method is simultaneously data-informed, exhibits provable guarantees on the size and performance of the pruned network, and is widely applicable to varying network architectures and data sets. Our analytical bounds bridge the notions of compressibility and importance of network structures, which gives rise to a fully-automated procedure for identifying and preserving filters in layers that are essential to the network's performance. Our experimental evaluations on popular architectures and data sets show that our algorithm consistently generates sparser and more efficient models than those constructed by existing filter pruning approaches.
Continuous-depth neural models, where the derivative of the model's hidden state is defined by a neural network, have enabled strong sequential data processing capabilities. However, these models rely on advanced numerical differential equation (DE) solvers resulting in a significant overhead both in terms of computational cost and model complexity. In this paper, we present a new family of models, termed Closed-form Continuous-depth (CfC) networks, that are simple to describe and at least one order of magnitude faster while exhibiting equally strong modeling abilities compared to their ODE-based counterparts. The models are hereby derived from the analytical closed-form solution of an expressive subset of time-continuous models, thus alleviating the need for complex DE solvers all together. In our experimental evaluations, we demonstrate that CfC networks outperform advanced, recurrent models over a diverse set of time-series prediction tasks, including those with long-term dependencies and irregularly sampled data. We believe our findings open new opportunities to train and deploy rich, continuous neural models in resource-constrained settings, which demand both performance and efficiency.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.