Title: Neuromechanical Autoencoders: Learning to Couple Elastic and Neural Network Nonlinearity
Intelligent biological systems are characterized by their embodiment in a complex environment and the intimate interplay between their nervous systems and the nonlinear mechanical properties of their bodies. This coordination, in which the dynamics of the motor system co-evolved to reduce the computational burden on the brain, is referred to as "mechanical intelligence" or "morphological computation". In this work, we seek to develop machine learning analogs of this process, in which we jointly learn the morphology of complex nonlinear elastic solids along with a deep neural network to control it. By using a specialized differentiable simulator of elastic mechanics coupled to conventional deep learning architectures---which we refer to as neuromechanical autoencoders---we are able to learn to perform morphological computation via gradient descent. Key to our approach is the use of mechanical metamaterials---cellular solids, in particular---as the morphological substrate. Just as deep neural networks provide flexible and massively-parametric function approximators for perceptual and control tasks, cellular solid metamaterials are promising as a rich and learnable space for approximating a variety of actuation tasks. In this work we take advantage of these complementary computational concepts to co-design materials and neural network controls to achieve nonintuitive mechanical behavior. We demonstrate in simulation how it is possible to achieve translation, rotation, and shape matching, as well as a "digital MNIST" task. We additionally manufacture and evaluate one of the designs to verify its real-world behavior. more »« less
Bhovad, Priyanka; Li, Suyi(
, Proc. SPIE 11589, Behavior and Mechanics of Multifunctional Materials XV)
null
(Ed.)
In the field of soft robotics, harnessing the nonlinear dynamics of soft and compliant bodies as a computational resource to enable embodied intelligence and control is known as morphological computation. Physical reservoir computing (PRC) is a true instance of morphological computation wherein; a physical nonlinear dynamic system is used as a fixed reservoir to perform complex computational tasks. These dynamic reservoirs can be used to approximate nonlinear dynamical systems and even perform machine learning tasks. By numerical simulation, this study illustrates that an origami meta-material can also be used as a dynamic reservoir for pattern generation, output modulation, and input sensing. These results could pave the way for intelligently designed origami-based robots that interact with the environment through a distributed network of sensors and actuators. This embodied intelligence will enable the next generations of soft robots to autonomously coordinate and modulate their activities, such as locomotion gait generation and limb manipulation while resisting external disturbances.
Bittner, Sean R; Palmigiano, Agostina; Piet, Alex T; Duan, Chunyu A; Brody, Carlos D; Miller, Kenneth D; Cunningham, John(
, eLife)
A cornerstone of theoretical neuroscience is the circuit model: a system of equations that captures a hypothesized neural mechanism. Such models are valuable when they give rise to an experimentally observed phenomenon -- whether behavioral or a pattern of neural activity -- and thus can offer insights into neural computation. The operation of these circuits, like all models, critically depends on the choice of model parameters. A key step is then to identify the model parameters consistent with observed phenomena: to solve the inverse problem. In this work, we present a novel technique, emergent property inference (EPI), that brings the modern probabilistic modeling toolkit to theoretical neuroscience. When theorizing circuit models, theoreticians predominantly focus on reproducing computational properties rather than a particular dataset. Our method uses deep neural networks to learn parameter distributions with these computational properties. This methodology is introduced through a motivational example of parameter inference in the stomatogastric ganglion. EPI is then shown to allow precise control over the behavior of inferred parameters and to scale in parameter dimension better than alternative techniques. In the remainder of this work, we present novel theoretical findings in models of primary visual cortex and superior colliculus, which were gained through the examination of complex parametric structure captured by EPI. Beyond its scientific contribution, this work illustrates the variety of analyses possible once deep learning is harnessed towards solving theoretical inverse problems.
Jean-Paul, S.; Elseify, T.; Obeid, I.; Picone, J.(
, IEEE Signal Processing in Medicine and Biology Symposium (SPMB))
Obeid, I.; Selesnik, I.; Picone, J.
(Ed.)
The Neuronix high-performance computing cluster allows us to conduct extensive machine learning experiments on big data [1]. This heterogeneous cluster uses innovative scheduling technology, Slurm [2], that manages a network of CPUs and graphics processing units (GPUs). The GPU farm consists of a variety of processors ranging from low-end consumer grade devices such as the Nvidia GTX 970 to higher-end devices such as the GeForce RTX 2080. These GPUs are essential to our research since they allow extremely compute-intensive deep learning tasks to be executed on massive data resources such as the TUH EEG Corpus [2]. We use TensorFlow [3] as the core machine learning library for our deep learning systems, and routinely employ multiple GPUs to accelerate the training process.
Reproducible results are essential to machine learning research. Reproducibility in this context means the ability to replicate an existing experiment – performance metrics such as error rates should be identical and floating-point calculations should match closely. Three examples of ways we typically expect an experiment to be replicable are: (1) The same job run on the same processor should produce the same results each time it is run. (2) A job run on a CPU and GPU should produce identical results. (3) A job should produce comparable results if the data is presented in a different order. System optimization requires an ability to directly compare error rates for algorithms evaluated under comparable operating conditions. However, it is a difficult task to exactly reproduce the results for large, complex deep learning systems that often require more than a trillion calculations per experiment [5]. This is a fairly well-known issue and one we will explore in this poster.
Researchers must be able to replicate results on a specific data set to establish the integrity of an implementation. They can then use that implementation as a baseline for comparison purposes. A lack of reproducibility makes it very difficult to debug algorithms and validate changes to the system. Equally important, since many results in deep learning research are dependent on the order in which the system is exposed to the data, the specific processors used, and even the order in which those processors are accessed, it becomes a challenging problem to compare two algorithms since each system must be individually optimized for a specific data set or processor. This is extremely time-consuming for algorithm research in which a single run often taxes a computing environment to its limits. Well-known techniques such as cross-validation [5,6] can be used to mitigate these effects, but this is also computationally expensive.
These issues are further compounded by the fact that most deep learning algorithms are susceptible to the way computational noise propagates through the system. GPUs are particularly notorious for this because, in a clustered environment, it becomes more difficult to control which processors are used at various points in time. Another equally frustrating issue is that upgrades to the deep learning package, such as the transition from TensorFlow v1.9 to v1.13, can also result in large fluctuations in error rates when re-running the same experiment. Since TensorFlow is constantly updating functions to support GPU use, maintaining an historical archive of experimental results that can be used to calibrate algorithm research is quite a challenge. This makes it very difficult to optimize the system or select the best configurations.
The overall impact of all of these issues described above is significant as error rates can fluctuate by as much as 25% due to these types of computational issues. Cross-validation is one technique used to mitigate this, but that is expensive since you need to do multiple runs over the data, which further taxes a computing infrastructure already running at max capacity.
GPUs are preferred when training a large network since these systems train at least two orders of magnitude faster than CPUs [7]. Large-scale experiments are simply not feasible without using GPUs. However, there is a tradeoff to gain this performance. Since all our GPUs use the NVIDIA CUDA® Deep Neural Network library (cuDNN) [8], a GPU-accelerated library of primitives for deep neural networks, it adds an element of randomness into the experiment. When a GPU is used to train a network in TensorFlow, it automatically searches for a cuDNN implementation. NVIDIA’s cuDNN implementation provides algorithms that increase the performance and help the model train quicker, but they are non-deterministic algorithms [9,10]. Since our networks have many complex layers, there is no easy way to avoid this randomness. Instead of comparing each epoch, we compare the average performance of the experiment because it gives us a hint of how our model is performing per experiment, and if the changes we make are efficient.
In this poster, we will discuss a variety of issues related to reproducibility and introduce ways we mitigate these effects. For example, TensorFlow uses a random number generator (RNG) which is not seeded by default. TensorFlow determines the initialization point and how certain functions execute using the RNG. The solution for this is seeding all the necessary components before training the model. This forces TensorFlow to use the same initialization point and sets how certain layers work (e.g., dropout layers). However, seeding all the RNGs will not guarantee a controlled experiment. Other variables can affect the outcome of the experiment such as training using GPUs, allowing multi-threading on CPUs, using certain layers, etc.
To mitigate our problems with reproducibility, we first make sure that the data is processed in the same order during training. Therefore, we save the data from the last experiment and to make sure the newer experiment follows the same order. If we allow the data to be shuffled, it can affect the performance due to how the model was exposed to the data. We also specify the float data type to be 32-bit since Python defaults to 64-bit. We try to avoid using 64-bit precision because the numbers produced by a GPU can vary significantly depending on the GPU architecture [11-13]. Controlling precision somewhat reduces differences due to computational noise even though technically it increases the amount of computational noise.
We are currently developing more advanced techniques for preserving the efficiency of our training process while also maintaining the ability to reproduce models. In our poster presentation we will demonstrate these issues using some novel visualization tools, present several examples of the extent to which these issues influence research results on electroencephalography (EEG) and digital pathology experiments and introduce new ways to manage such computational issues.
Huang, Jiayi; Majumder, Pritam; Kim, Sungkeun; Muzahid, Abdullah; Yum, Ki Hwan; Kim, Eun Jung(
, 2021 ACM/IEEE 48th Annual International Symposium on Computer Architecture (ISCA))
Large-scale distributed deep learning training has enabled developments of more complex deep neural network models to learn from larger datasets for sophisticated tasks. In particular, distributed stochastic gradient descent intensively invokes all-reduce operations for gradient update, which dominates communication time during iterative training epochs. In this work, we identify the inefficiency in widely used all-reduce algorithms, and the opportunity of algorithm-architecture co-design. We propose MultiTree all-reduce algorithm with topology and resource utilization awareness for efficient and scalable all-reduce operations, which is applicable to different interconnect topologies. Moreover, we co-design the network interface to schedule and coordinate the all-reduce messages for contention-free communications, working in synergy with the algorithm. The flow control is also simplified to exploit the bulk data transfer of big gradient exchange. We evaluate the co-design using different all-reduce data sizes for synthetic study, demonstrating its effectiveness on various interconnection network topologies, in addition to state-of-the-art deep neural networks for real workload experiments. The results show that MultiTree achieves 2.3× and 1.56× communication speedup, as well as up to 81% and 30% training time reduction compared to ring all-reduce and state-of-the-art approaches, respectively.
Wu, Tailin; Maruyama, Takashi; Zhao, Qingqing; Wetzstein, Gordon; Leskovec, Jure(
, International Conference on Learning Representations (ICLR))
Simulating the time evolution of physical systems is pivotal in many scientific and engineering problems. An open challenge in simulating such systems is their multi-resolution dynamics: a small fraction of the system is extremely dynamic, and requires very fine-grained resolution, while a majority of the system is changing slowly and can be modeled by coarser spatial scales. Typical learning-based surrogate models use a uniform spatial scale, which needs to resolve to the finest required scale and can waste a huge compute to achieve required accuracy. We introduced Learning controllable Adaptive simulation for Multiresolution Physics (LAMP) as the first full deep learning-based surrogate model that jointly learns the evolution model and optimizes appropriate spatial resolutions that devote more compute to the highly dynamic regions. LAMP consists of a Graph Neural Network (GNN) for learning the forward evolution, and a GNNbased actor-critic for learning the policy of spatial refinement and coarsening. We introduced learning techniques that optimize LAMP with weighted sum of error and computational cost as objective, allowing LAMP to adapt to varying relative importance of error vs. computation tradeoff at inference time. We evaluated our method in a 1D benchmark of nonlinear PDEs and a challenging 2D mesh-based simulation. We demonstrated that our LAMP outperforms state-of-the-art deep learning surrogate models, and can adaptively trade-off computation to improve long-term prediction error: it achieves an average of 33.7% error reduction for 1D nonlinear PDEs, and outperforms MeshGraphNets + classical Adaptive Mesh Refinement (AMR) in 2D mesh-based simulations.
Oktay, Deniz, Mirramezani, Mehran, Medina, Eder, and Adams, Ryan P. Neuromechanical Autoencoders: Learning to Couple Elastic and Neural Network Nonlinearity. Retrieved from https://par.nsf.gov/biblio/10470418.
Oktay, Deniz, Mirramezani, Mehran, Medina, Eder, & Adams, Ryan P. Neuromechanical Autoencoders: Learning to Couple Elastic and Neural Network Nonlinearity. Retrieved from https://par.nsf.gov/biblio/10470418.
Oktay, Deniz, Mirramezani, Mehran, Medina, Eder, and Adams, Ryan P.
"Neuromechanical Autoencoders: Learning to Couple Elastic and Neural Network Nonlinearity". Country unknown/Code not available: International Conference on Learning Representations (ICLR). https://par.nsf.gov/biblio/10470418.
@article{osti_10470418,
place = {Country unknown/Code not available},
title = {Neuromechanical Autoencoders: Learning to Couple Elastic and Neural Network Nonlinearity},
url = {https://par.nsf.gov/biblio/10470418},
abstractNote = {Intelligent biological systems are characterized by their embodiment in a complex environment and the intimate interplay between their nervous systems and the nonlinear mechanical properties of their bodies. This coordination, in which the dynamics of the motor system co-evolved to reduce the computational burden on the brain, is referred to as "mechanical intelligence" or "morphological computation". In this work, we seek to develop machine learning analogs of this process, in which we jointly learn the morphology of complex nonlinear elastic solids along with a deep neural network to control it. By using a specialized differentiable simulator of elastic mechanics coupled to conventional deep learning architectures---which we refer to as neuromechanical autoencoders---we are able to learn to perform morphological computation via gradient descent. Key to our approach is the use of mechanical metamaterials---cellular solids, in particular---as the morphological substrate. Just as deep neural networks provide flexible and massively-parametric function approximators for perceptual and control tasks, cellular solid metamaterials are promising as a rich and learnable space for approximating a variety of actuation tasks. In this work we take advantage of these complementary computational concepts to co-design materials and neural network controls to achieve nonintuitive mechanical behavior. We demonstrate in simulation how it is possible to achieve translation, rotation, and shape matching, as well as a "digital MNIST" task. We additionally manufacture and evaluate one of the designs to verify its real-world behavior.},
journal = {},
publisher = {International Conference on Learning Representations (ICLR)},
author = {Oktay, Deniz and Mirramezani, Mehran and Medina, Eder and Adams, Ryan P.},
}
Warning: Leaving National Science Foundation Website
You are now leaving the National Science Foundation website to go to a non-government website.
Website:
NSF takes no responsibility for and exercises no control over the views expressed or the accuracy of
the information contained on this site. Also be aware that NSF's privacy policy does not apply to this site.
