scholarly journals Any-Precision Deep Neural Networks

2020 ◽  
Author(s):  
Haichao Yu ◽  
Haoxiang Li ◽  
Honghui Shi ◽  
Thomas S. Huang ◽  
Gang Hua
Keyword(s):  
Neural Networks ◽  
Deep Neural Networks ◽  
Nice Property ◽  
Model Accuracy ◽  
Trade Off ◽  
Trade Offs ◽  
Real World Applications ◽  
Numerical Precision ◽  
Speed And Accuracy ◽  
Least Significant Bits

We present Any-Precision Deep Neural Networks (Any- Precision DNNs), which are trained with a new method that empowers learned DNNs to be flexible in any numerical precision during inference. The same model in runtime can be flexibly and directly set to different bit-width, by trun- cating the least significant bits, to support dynamic speed and accuracy trade-off. When all layers are set to low- bits, we show that the model achieved accuracy compara- ble to dedicated models trained at the same precision. This nice property facilitates flexible deployment of deep learn- ing models in real-world applications, where in practice trade-offs between model accuracy and runtime efficiency are often sought. Previous literature presents solutions to train models at each individual fixed efficiency/accuracy trade-off point. But how to produce a model flexible in runtime precision is largely unexplored. When the demand of efficiency/accuracy trade-off varies from time to time or even dynamically changes in runtime, it is infeasible to re-train models accordingly, and the storage budget may forbid keeping multiple models. Our proposed framework achieves this flexibility without performance degradation. More importantly, we demonstrate that this achievement is agnostic to model architectures. We experimentally validated our method with different deep network backbones (AlexNet-small, Resnet-20, Resnet-50) on different datasets (SVHN, Cifar-10, ImageNet) and observed consistent results.

Dual Goals for Speed and Accuracy on the Same Performance Task

2012 ◽  
Vol 11 (3) ◽  
pp. 118-126 ◽  
Author(s):  
Olive Emil Wetter ◽  
Jürgen Wegge ◽  
Klaus Jonas ◽  
Klaus-Helmut Schmidt
Keyword(s):  
Memory Scanning ◽  
Performance Task ◽  
Performance Tasks ◽  
Trade Off ◽  
Test Experiment ◽  
Trade Offs ◽  
New Finding ◽  
Sternberg Paradigm ◽  
Speed Accuracy ◽  
Speed And Accuracy

In most work contexts, several performance goals coexist, and conflicts between them and trade-offs can occur. Our paper is the first to contrast a dual goal for speed and accuracy with a single goal for speed on the same task. The Sternberg paradigm (Experiment 1, n = 57) and the d2 test (Experiment 2, n = 19) were used as performance tasks. Speed measures and errors revealed in both experiments that dual as well as single goals increase performance by enhancing memory scanning. However, the single speed goal triggered a speed-accuracy trade-off, favoring speed over accuracy, whereas this was not the case with the dual goal. In difficult trials, dual goals slowed down scanning processes again so that errors could be prevented. This new finding is particularly relevant for security domains, where both aspects have to be managed simultaneously.

Chapter 15. Human-Centered Concept Explanations for Neural Networks

2021 ◽  
Author(s):  
Chih-Kuan Yeh ◽  
Been Kim ◽  
Pradeep Ravikumar
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Case Studies ◽  
Real World ◽  
Deep Neural Networks ◽  
Learning Models ◽  
Real World Applications ◽  
The Right ◽  
Concept Activation ◽  
Machine Learning Models

Understanding complex machine learning models such as deep neural networks with explanations is crucial in various applications. Many explanations stem from the model perspective, and may not necessarily effectively communicate why the model is making its predictions at the right level of abstraction. For example, providing importance weights to individual pixels in an image can only express which parts of that particular image is important to the model, but humans may prefer an explanation which explains the prediction by concept-based thinking. In this work, we review the emerging area of concept based explanations. We start by introducing concept explanations including the class of Concept Activation Vectors (CAV) which characterize concepts using vectors in appropriate spaces of neural activations, and discuss different properties of useful concepts, and approaches to measure the usefulness of concept vectors. We then discuss approaches to automatically extract concepts, and approaches to address some of their caveats. Finally, we discuss some case studies that showcase the utility of such concept-based explanations in synthetic settings and real world applications.

scholarly journals Transfer Learning and Deep Domain Adaptation

2020 ◽  
Author(s):  
Wen Xu ◽  
Jing He ◽  
Yanfeng Shu
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Transfer Learning ◽  
Real World ◽  
Deep Neural Networks ◽  
Domain Adaptation ◽  
Fine Tuning ◽  
Real World Applications ◽  
Comprehensive Survey ◽  
Sample Reconstruction

Transfer learning is an emerging technique in machine learning, by which we can solve a new task with the knowledge obtained from an old task in order to address the lack of labeled data. In particular deep domain adaptation (a branch of transfer learning) gets the most attention in recently published articles. The intuition behind this is that deep neural networks usually have a large capacity to learn representation from one dataset and part of the information can be further used for a new task. In this research, we firstly present the complete scenarios of transfer learning according to the domains and tasks. Secondly, we conduct a comprehensive survey related to deep domain adaptation and categorize the recent advances into three types based on implementing approaches: fine-tuning networks, adversarial domain adaptation, and sample-reconstruction approaches. Thirdly, we discuss the details of these methods and introduce some typical real-world applications. Finally, we conclude our work and explore some potential issues to be further addressed.

scholarly journals Optimizing for Interpretability in Deep Neural Networks with Tree Regularization

2021 ◽  
Vol 72 ◽  
pp. 1-37
Author(s):  
Mike Wu ◽  
Sonali Parbhoo ◽  
Michael C. Hughes ◽  
Volker Roth ◽  
Finale Doshi-Velez
Keyword(s):  
Neural Networks ◽  
Predictive Power ◽  
Deep Neural Networks ◽  
Large Body ◽  
Black Box ◽  
New Family ◽  
Deep Model ◽  
Real World Applications ◽  
Adversarial Examples ◽  
Key Barrier

Deep models have advanced prediction in many domains, but their lack of interpretability  remains a key barrier to the adoption in many real world applications. There exists a large  body of work aiming to help humans understand these black box functions to varying levels  of granularity – for example, through distillation, gradients, or adversarial examples. These  methods however, all tackle interpretability as a separate process after training. In this  work, we take a different approach and explicitly regularize deep models so that they are  well-approximated by processes that humans can step through in little time. Specifically,  we train several families of deep neural networks to resemble compact, axis-aligned decision  trees without significant compromises in accuracy. The resulting axis-aligned decision  functions uniquely make tree regularized models easy for humans to interpret. Moreover,  for situations in which a single, global tree is a poor estimator, we introduce a regional tree regularizer that encourages the deep model to resemble a compact, axis-aligned decision  tree in predefined, human-interpretable contexts. Using intuitive toy examples, benchmark  image datasets, and medical tasks for patients in critical care and with HIV, we demonstrate  that this new family of tree regularizers yield models that are easier for humans to simulate  than L1 or L2 penalties without sacrificing predictive power. 

scholarly journals Towards Understanding the Risks of Gradient Inversion in Federated Learning

2021 ◽  
Author(s):  
Ali Hatamizadeh ◽  
Hongxu Yin ◽  
Pavlo Molchanov ◽  
Andriy Myronenko ◽  
Wenqi Li ◽  
...  
Keyword(s):  
Neural Networks ◽  
Data Privacy ◽  
Deep Neural Networks ◽  
Differential Privacy ◽  
Use Cases ◽  
Training Data ◽  
Model Accuracy ◽  
Healthcare Applications ◽  
Raw Data ◽  
Collaborative Training

Abstract Federated learning (FL) allows the collaborative training of AI models without needing to share raw data. This capability makes it especially interesting for healthcare applications where patient and data privacy is of utmost concern. However, recent works on the inversion of deep neural networks from model gradients raised concerns about the security of FL in preventing the leakage of training data. In this work, we show that these attacks presented in the literature are impractical in real FL use-cases and provide a new baseline attack that works for more realistic scenarios where the clients’ training involves updating the Batch Normalization (BN) statistics. Furthermore, we present new ways to measure and visualize potential data leakage in FL. Our work is a step towards establishing reproducible methods of measuring data leakage in FL and could help determine the optimal tradeoffs between privacy-preserving techniques, such as differential privacy, and model accuracy based on quantifiable metrics.

scholarly journals Bias in Deep Neural Networks in Land Use Characterization for International Development

2021 ◽  
Vol 13 (15) ◽  
pp. 2908
Author(s):  
Do-Hyung Kim ◽  
Guzmán López ◽  
Diego Kiedanski ◽  
Iyke Maduako ◽  
Braulio Ríos ◽  
...  
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Land Use ◽  
International Development ◽  
Deep Neural Networks ◽  
Satellite Image ◽  
Fine Tuning ◽  
Training Dataset ◽  
Model Accuracy ◽  
The Neural Network

Understanding the biases in Deep Neural Networks (DNN) based algorithms is gaining paramount importance due to its increased applications on many real-world problems. A known problem of DNN penalizing the underrepresented population could undermine the efficacy of development projects dependent on data produced using DNN-based models. In spite of this, the problems of biases in DNN for Land Use and Land Cover Classification (LULCC) have not been a subject of many studies. In this study, we explore ways to quantify biases in DNN for land use with an example of identifying school buildings in Colombia from satellite imagery. We implement a DNN-based model by fine-tuning an existing, pre-trained model for school building identification. The model achieved overall 84% accuracy. Then, we used socioeconomic covariates to analyze possible biases in the learned representation. The retrained deep neural network was used to extract visual features (embeddings) from satellite image tiles. The embeddings were clustered into four subtypes of schools, and the accuracy of the neural network model was assessed for each cluster. The distributions of various socioeconomic covariates by clusters were analyzed to identify the links between the model accuracy and the aforementioned covariates. Our results indicate that the model accuracy is lowest (57%) where the characteristics of the landscape are predominantly related to poverty and remoteness, which confirms our original assumption on the heterogeneous performances of Artificial Intelligence (AI) algorithms and their biases. Based on our findings, we identify possible sources of bias and present suggestions on how to prepare a balanced training dataset that would result in less biased AI algorithms. The framework used in our study to better understand biases in DNN models would be useful when Machine Learning (ML) techniques are adopted in lieu of ground-based data collection for international development programs. Because such programs aim to solve issues of social inequality, MLs are only applicable when they are transparent and accountable.

scholarly journals Survey of deployment locations and underlying hardware architectures for contemporary deep neural networks

2019 ◽  
Vol 15 (8) ◽  
pp. 155014771986866
Author(s):  
Miloš Kotlar ◽  
Dragan Bojić ◽  
Marija Punt ◽  
Veljko Milutinović
Keyword(s):  
Neural Networks ◽  
Integrated Circuits ◽  
Data Processing ◽  
Deep Neural Networks ◽  
Multicore Processors ◽  
Two Dimensions ◽  
Processing Unit ◽  
Central Processing ◽  
Trade Offs ◽  
Dew Computing

This article overviews the emerging use of deep neural networks in data analytics and explores which type of underlying hardware and architectural approach is best used in various deployment locations when implementing deep neural networks. The locations which are discussed are in the cloud, fog, and dew computing (dew computing is performed by end devices). Covered architectural approaches include multicore processors (central processing unit), manycore processors (graphics processing unit), field programmable gate arrays, and application-specific integrated circuits. The proposed classification in this article divides the existing solutions into 12 different categories, organized in two dimensions. The proposed classification allows a comparison of existing architectures, which are predominantly cloud-based, and anticipated future architectures, which are expected to be hybrid cloud-fog-dew architectures for applications in Internet of Things and Wireless Sensor Networks. Researchers interested in studying trade-offs among data processing bandwidth, data processing latency, and processing power consumption would benefit from the classification made in this article.

scholarly journals Spiking Neural Networks for Computational Intelligence: An Overview

2021 ◽  
Vol 5 (4) ◽  
pp. 67
Author(s):  
Shirin Dora ◽  
Nikola Kasabov
Keyword(s):  
Neural Networks ◽  
Energy Efficiency ◽  
Computational Intelligence ◽  
Deep Neural Networks ◽  
Resource Constraints ◽  
Spiking Neural Networks ◽  
Low Latency ◽  
Temporal Data ◽  
Tremendous Progress ◽  
Real World Applications

Deep neural networks with rate-based neurons have exhibited tremendous progress in the last decade. However, the same level of progress has not been observed in research on spiking neural networks (SNN), despite their capability to handle temporal data, energy-efficiency and low latency. This could be because the benchmarking techniques for SNNs are based on the methods used for evaluating deep neural networks, which do not provide a clear evaluation of the capabilities of SNNs. Particularly, the benchmarking of SNN approaches with regards to energy efficiency and latency requires realization in suitable hardware, which imposes additional temporal and resource constraints upon ongoing projects. This review aims to provide an overview of the current real-world applications of SNNs and identifies steps to accelerate research involving SNNs in the future.

scholarly journals Energy-efficient Amortized Inference with Cascaded Deep Classifiers

2018 ◽  
Author(s):  
Jiaqi Guan ◽  
Yang Liu ◽  
Qiang Liu ◽  
Jian Peng
Keyword(s):  
Neural Networks ◽  
Energy Cost ◽  
Deep Neural Networks ◽  
Predictive Accuracy ◽  
Computational Cost ◽  
Mobile Sensing ◽  
Test Time ◽  
Trade Off ◽  
Effective Cost ◽  
Energy Constrained

Deep neural networks have been remarkable successful in various AI tasks but often cast high computation and energy cost for energy-constrained applications such as mobile sensing. We address this problem by proposing a novel framework that optimizes the prediction accuracy and energy cost simultaneously, thus enabling effective cost-accuracy trade-off at test time. In our framework, each data instance is pushed into a cascade of deep neural networks with increasing sizes, and a selection module is used to sequentially determine when a sufficiently accurate classifier can be used for this data instance. The cascade of neural networks and the selection module are jointly trained in an end-to-end fashion by the REINFORCE algorithm to optimize a trade-off between the computational cost and the predictive accuracy. Our method is able to simultaneously improve the accuracy and efficiency by learning to assign easy instances to fast yet sufficiently accurate classifiers to save computation and energy cost, while assigning harder instances to deeper and more powerful classifiers to ensure satisfiable accuracy. Moreover, we demonstrate our method's effectiveness with extensive experiments on CIFAR-10/100, ImageNet32x32 and original ImageNet dataset.

Sign in / Sign up

Export Citation Format

Share Document