Abstract: "What if a team of collaborative autonomous robots grew your food for you? In this talk, I will discuss some key advances in robotics, machine learning, and autonomy that will one day enable teams of small robots to grow food for you in your backyard in a fundamentally more sustainable way than modern mega-farms! Teams of small aerial and ground robots could be a potential solution to many of the serious problems that modern agriculture is facing. However, fully autonomous robots that operate without supervision for weeks, months, or entire growing season are not yet practical. I will discuss my group's theoretical and practical work towards the underlying challenging problems in robotic systems, autonomy, sensing, and learning. I will begin with our lightweight, compact, and autonomous field robot TerraSentia and the recent successes of this type of undercanopy robots for high-throughput phenotyping with deep learning-based machine vision. I will also discuss how to make a team of autonomous robots learn to coordinate to weed large agricultural farms under partial observability. These direct applications will help me make the case for the type of reinforcement learning and adaptive control that are necessary to usher in the next generation of autonomous field robots that learn to solve complex problems in harsh, changing, and dynamic environments. I will then end with an overview of our new MURI, in which we are working towards developing AI and control that leverages neurodynamics inspired by the Octopus brain."

As Deep Learning (DL) models have been increasingly used in latency-sensitive applications, there has been a growing interest in improving their response time. An important venue for such improvement is to profile the execution of these models and characterize their performance to identify possible optimization opportunities. However, the current profiling tools lack the highly desired abilities to characterize ideal performance, identify sources of inefficiency, and quantify the benefits of potential optimizations. Such deficiencies have led to slow characterization/optimization cycles that cannot keep up with the fast pace at which new DL models are introduced. We propose Benanza, a sustainable and extensible benchmarking and analysis design that speeds up the characterization/optimization cycle of DL models on GPUs. Benanza consists of four major components: a model processor that parses models into an internal representation, a configurable benchmark generator that automatically generates micro-benchmarks given a set of models, a database of benchmark results, and an analyzer that computes the "lower-bound" latency of DL models using the benchmark data and informs optimizations of model execution. The "lower-bound" latency metric estimates the ideal model execution on a GPU system and serves as the basis for identifying optimization opportunities in frameworks or system libraries. We used Benanza to evaluate 30 ONNX models in MXNet, ONNX Runtime, and PyTorch on 7 GPUs ranging from Kepler to the latest Turing, and identified optimizations in parallel layer execution, cuDNN convolution algorithm selection, framework inefficiency, layer fusion, and using Tensor Cores.

The current Deep Learning (DL) landscape is fast-paced and is rife with non-uniform models, hardware/software (HW/SW) stacks, but lacks a DL benchmarking platform to facilitate evaluation and comparison of DL innovations, be it models, frameworks, libraries, or hardware. Due to the lack of a benchmarking platform, the current practice of evaluating the benefits of proposed DL innovations is both arduous and error-prone - stifling the adoption of the innovations. In this work, we first identify $10$ design features which are desirable within a DL benchmarking platform. These features include: performing the evaluation in a consistent, reproducible, and scalable manner, being framework and hardware agnostic, supporting real-world benchmarking workloads, providing in-depth model execution inspection across the HW/SW stack levels, etc. We then propose MLModelScope, a DL benchmarking platform design that realizes the $10$ objectives. MLModelScope proposes a specification to define DL model evaluations and techniques to provision the evaluation workflow using the user-specified HW/SW stack. MLModelScope defines abstractions for frameworks and supports board range of DL models and evaluation scenarios. We implement MLModelScope as an open-source project with support for all major frameworks and hardware architectures. Through MLModelScope's evaluation and automated analysis workflows, we performed case-study analyses of $37$ models across $4$ systems and show how model, hardware, and framework selection affects model accuracy and performance under different benchmarking scenarios. We further demonstrated how MLModelScope's tracing capability gives a holistic view of model execution and helps pinpoint bottlenecks.

The understanding of the dynamics of the velocity gradients in turbulent flows is critical to understanding various non-linear turbulent processes. The pressure-Hessian and the viscous-Laplacian govern the evolution of the velocity-gradients and are known to be non-local in nature. Over the years, several simplified dynamical models have been proposed that models the viscous-Laplacian and the pressure-Hessian primarily in terms of local velocity gradients information. These models can also serve as closure models for the Lagrangian PDF methods. The recent fluid deformation closure model (RFDM) has been shown to retrieve excellent one-time statistics of the viscous process. However, the pressure-Hessian modelled by the RFDM has various physical limitations. In this work, we first demonstrate the limitations of the RFDM in estimating the pressure-Hessian. Further, we employ a tensor basis neural network (TBNN) to model the pressure-Hessian from the velocity gradient tensor itself. The neural network is trained on high-resolution data obtained from direct numerical simulation (DNS) of isotropic turbulence at Reynolds number of 433 (JHU turbulence database, JHTD). The predictions made by the TBNN are tested against two different isotropic turbulence datasets at Reynolds number of 433 (JHTD) and 315 (UP Madrid turbulence database, UPMTD) and channel flow dataset at Reynolds number of 1000 (UT Texas and JHTD). The evaluation of the neural network output is made in terms of the alignment statistics of the predicted pressure-Hessian eigenvectors with the strain-rate eigenvectors for turbulent isotropic flow as well as channel flow. Our analysis of the predicted solution leads to the discovery of ten unique coefficients of the tensor basis of strain-rate and rotation-rate tensors, the linear combination over which accurately captures key alignment statistics of the pressure-Hessian tensor.

Transfer learning have been frequently used to improve deep neural network training through incorporating weights of pre-trained networks as the starting-point of optimization for regularization. While deep transfer learning can usually boost the performance with better accuracy and faster convergence, transferring weights from inappropriate networks hurts training procedure and may lead to even lower accuracy. In this paper, we consider deep transfer learning as minimizing a linear combination of empirical loss and regularizer based on pre-trained weights, where the regularizer would restrict the training procedure from lowering the empirical loss, with conflicted descent directions (e.g., derivatives). Following the view, we propose a novel strategy making regularization-based Deep Transfer learning Never Hurt (DTNH) that, for each iteration of training procedure, computes the derivatives of the two terms separately, then re-estimates a new descent direction that does not hurt the empirical loss minimization while preserving the regularization affects from the pre-trained weights. Extensive experiments have been done using common transfer learning regularizers, such as L2-SP and knowledge distillation, on top of a wide range of deep transfer learning benchmarks including Caltech, MIT indoor 67, CIFAR-10 and ImageNet. The empirical results show that the proposed descent direction estimation strategy DTNH can always improve the performance of deep transfer learning tasks based on all above regularizers, even when transferring pre-trained weights from inappropriate networks. All in all, DTNH strategy can improve state-of-the-art regularizers in all cases with 0.1%--7% higher accuracy in all experiments.

Deep learning approaches have recently achieved impressive performance on both audio source separation and sound classification. Most audio source separation approaches focus only on separating sources belonging to a restricted domain of source classes, such as speech and music. However, recent work has demonstrated the possibility of "universal sound separation", which aims to separate acoustic sources from an open domain, regardless of their class. In this paper, we utilize the semantic information learned by sound classifier networks trained on a vast amount of diverse sounds to improve universal sound separation. In particular, we show that semantic embeddings extracted from a sound classifier can be used to condition a separation network, providing it with useful additional information. This approach is especially useful in an iterative setup, where source estimates from an initial separation stage and their corresponding classifier-derived embeddings are fed to a second separation network. By performing a thorough hyperparameter search consisting of over a thousand experiments, we find that classifier embeddings from clean sources provide nearly one dB of SNR gain, and our best iterative models achieve a significant fraction of this oracle performance, establishing a new state-of-the-art for universal sound separation.

--Existing approaches for automatically generating mathematical word problems are deprived of customizability and creativity due to the inherent nature of template-based mechanisms they employ. We present a solution to this problem with the use of deep neural language generation mechanisms. Our approach uses a Character Level Long Short T erm Memory Network (LSTM) to generate word problems, and uses POS (Part of Speech) tags to resolve the constraints found in the generated problems. Our approach is capable of generating Mathematics Word Problems in both English and Sinhala languages with an accuracy over 90%. A Mathematical word problem (MWP) is a mathematical problem expressed in natural language. Unlike other knowledge based question types such as travel or history related questions, MWPs require problem solving ability. In particular, algebraic questions involve sentences to make the questions more deep and inspective. Algebra is a major component of mathematics that is learnt by every student in Ordinary Level (O/L). Simple algebra problems mostly appear in a word format.

--Named entity recognition (NER) plays an important role in text-based information retrieval. In this paper, we combine Bidirectional Long Short-T erm Memory (Bi-LSTM) [7], [27] with Conditional Random Field (CRF) [9] to create a novel deep learning model for the NER problem. Each word as input of the deep learning model is represented by a Word2vec-trained vector . A word embedding set trained from about one million articles in 2018 collected through a Vietnamese news portal (baomoi.com). In addition, we concatenate a Word2V ec [18]- trained vector with semantic feature vector (Part-Of-Speech (POS) tagging, chunk-tag) and hidden syntactic feature vector (extracted by Bi-LSTM nerwork) to achieve the (so far best) result in Vietnamese NER system.

In this work we perform a study of various unsupervised methods to identify mental stress in firefighter trainees based on unlabeled heart rate variability data. We collect RR interval time series data from nearly 100 firefighter trainees that participated in a drill. We explore and compare three methods in order to perform unsupervised stress detection: 1) traditional K-Means clustering with engineered time and frequency domain features 2) convolutional autoencoders and 3) long short-term memory (LSTM) autoencoders, both trained on the raw RRI measurements combined with DBSCAN clustering and K-Nearest-Neighbors classification. We demonstrate that K-Means combined with engineered features is unable to capture meaningful structure within the data. On the other hand, convolutional and LSTM autoencoders tend to extract varying structure from the data pointing to different clusters with different sizes of clusters. We attempt at identifying the true stressed and normal clusters using the HRV markers of mental stress reported in the literature. We demonstrate that the clusters produced by the convolutional autoencoders consistently and successfully stratify stressed versus normal samples, as validated by several established physiological stress markers such as RMSSD, Max-HR, Mean-HR and LF-HF ratio.

Machine learning models are vulnerable to adversarial inputs that induce seemingly unjustifiable errors. As automated classifiers are increasingly used in industrial control systems and machinery, these adversarial errors could grow to be a serious problem. Despite numerous studies over the past few years, the field of adversarial ML is still considered alchemy, with no practical unbroken defenses demonstrated to date, leaving PHM practitioners with few meaningful ways of addressing the problem. We introduce turbidity detection as a practical superset of the adversarial input detection problem, coping with adversarial campaigns rather than statistically invisible one-offs. This perspective is coupled with ROCtheoretic design guidance that prescribes an inexpensive domain adaptation layer at the output of a deep learning model during an attack campaign. The result aims to approximate the Bayes optimal mitigation that ameliorates the detection model's degraded health. A proactively reactive type of prognostics is achieved via Monte Carlo simulation of various adversarial campaign scenarios, by sampling from the model's own turbidity distribution to quickly deploy the correct mitigation during a real-world campaign. A machine learning application often begins with a dataset of examples and the task is to find a classification model that will turn inputs into class-label predictions, while preserving some sense of minimum expected error. This is an open-access article distributed under the terms of the Creative Commons Attribution 3.0 United States License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. But less obviously, it is often possible to determin-istically find input examples that force the model to misclas-sify (Szegedy et al., 2014).