Machine Learning

---

GCNet: Non-local Networks Meet Squeeze-Excitation Networks and Beyond (1904.11492v1)

Yue Cao, Jiarui Xu, Stephen Lin, Fangyun Wei, Han Hu

2019-04-25

The Non-Local Network (NLNet) presents a pioneering approach for capturing long-range dependencies, via aggregating query-specific global context to each query position. However, through a rigorous empirical analysis, we have found that the global contexts modeled by non-local network are almost the same for different query positions within an image. In this paper, we take advantage of this finding to create a simplified network based on a query-independent formulation, which maintains the accuracy of NLNet but with significantly less computation. We further observe that this simplified design shares similar structure with Squeeze-Excitation Network (SENet). Hence we unify them into a three-step general framework for global context modeling. Within the general framework, we design a better instantiation, called the global context (GC) block, which is lightweight and can effectively model the global context. The lightweight property allows us to apply it for multiple layers in a backbone network to construct a global context network (GCNet), which generally outperforms both simplified NLNet and SENet on major benchmarks for various recognition tasks. The code and configurations are released at https://github.com/xvjiarui/GCNet.

Local Relation Networks for Image Recognition (1904.11491v1)

Han Hu, Zheng Zhang, Zhenda Xie, Stephen Lin

2019-04-25

The convolution layer has been the dominant feature extractor in computer vision for years. However, the spatial aggregation in convolution is basically a pattern matching process that applies fixed filters which are inefficient at modeling visual elements with varying spatial distributions. This paper presents a new image feature extractor, called the local relation layer, that adaptively determines aggregation weights based on the compositional relationship of local pixel pairs. With this relational approach, it can composite visual elements into higher-level entities in a more efficient manner that benefits semantic inference. A network built with local relation layers, called the Local Relation Network (LR-Net), is found to provide greater modeling capacity than its counterpart built with regular convolution on large-scale recognition tasks such as ImageNet classification.

Making Convolutional Networks Shift-Invariant Again (1904.11486v1)

Richard Zhang

2019-04-25

Modern convolutional networks are not shift-invariant, as small input shifts or translations can cause drastic changes in the output. Commonly used downsampling methods, such as max-pooling, strided-convolution, and average-pooling, ignore the sampling theorem. The well-known signal processing fix is anti-aliasing by low-pass filtering before downsampling. However, simply inserting this module into deep networks leads to performance degradation; as a result, it is seldomly used today. We show that when integrated correctly, it is compatible with existing architectural components, such as max-pooling. The technique is general and can be incorporated across layer types and applications, such as image classification and conditional image generation. In addition to increased shift-invariance, we also observe, surprisingly, that anti-aliasing boosts accuracy in ImageNet classification, across several commonly-used architectures. This indicates that anti-aliasing serves as effective regularization. Our results demonstrate that this classical signal processing technique has been undeservingly overlooked in modern deep networks. Code and anti-aliased versions of popular networks will be made available at \url{https://richzhang.github.io/antialiased-cnns/} .

Safe Reinforcement Learning with Scene Decomposition for Navigating Complex Urban Environments (1904.11483v1)

Maxime Bouton, Alireza Nakhaei, Kikuo Fujimura, Mykel J. Kochenderfer

2019-04-25

Navigating urban environments represents a complex task for automated vehicles. They must reach their goal safely and efficiently while considering a multitude of traffic participants. We propose a modular decision making algorithm to autonomously navigate intersections, addressing challenges of existing rule-based and reinforcement learning (RL) approaches. We first present a safe RL algorithm relying on a model-checker to ensure safety guarantees. To make the decision strategy robust to perception errors and occlusions, we introduce a belief update technique using a learning based approach. Finally, we use a scene decomposition approach to scale our algorithm to environments with multiple traffic participants. We empirically demonstrate that our algorithm outperforms rule-based methods and reinforcement learning techniques on a complex intersection scenario.

Importance of Copying Mechanism for News Headline Generation (1904.11475v1)

Ilya Gusev

2019-04-25

News headline generation is an essential problem of text summarization because it is constrained, well-defined, and is still hard to solve. Models with a limited vocabulary can not solve it well, as new named entities can appear regularly in the news and these entities often should be in the headline. News articles in morphologically rich languages such as Russian require model modifications due to a large number of possible word forms. This study aims to validate that models with a possibility of copying words from the original article performs better than models without such an option. The proposed model achieves a mean ROUGE score of 23 on the provided test dataset, which is 8 points greater than the result of a similar model without a copying mechanism. Moreover, the resulting model performs better than any known model on the new dataset of Russian news.

Sensor Fusion for Joint 3D Object Detection and Semantic Segmentation (1904.11466v1)

Gregory P. Meyer, Jake Charland, Darshan Hegde, Ankit Laddha, Carlos Vallespi-Gonzalez

2019-04-25

In this paper, we present an extension to LaserNet, an efficient and state-of-the-art LiDAR based 3D object detector. We propose a method for fusing image data with the LiDAR data and show that this sensor fusion method improves the detection performance of the model especially at long ranges. The addition of image data is straightforward and does not require image labels. Furthermore, we expand the capabilities of the model to perform 3D semantic segmentation in addition to 3D object detection. On a large benchmark dataset, we demonstrate our approach achieves state-of-the-art performance on both object detection and semantic segmentation while maintaining a low runtime.

Ray Interference: a Source of Plateaus in Deep Reinforcement Learning (1904.11455v1)

Tom Schaul, Diana Borsa, Joseph Modayil, Razvan Pascanu

2019-04-25

Rather than proposing a new method, this paper investigates an issue present in existing learning algorithms. We study the learning dynamics of reinforcement learning (RL), specifically a characteristic coupling between learning and data generation that arises because RL agents control their future data distribution. In the presence of function approximation, this coupling can lead to a problematic type of 'ray interference', characterized by learning dynamics that sequentially traverse a number of performance plateaus, effectively constraining the agent to learn one thing at a time even when learning in parallel is better. We establish the conditions under which ray interference occurs, show its relation to saddle points and obtain the exact learning dynamics in a restricted setting. We characterize a number of its properties and discuss possible remedies.

Faster and More Accurate Learning with Meta Trace Adaptation (1904.11439v1)

Mingde Zhao, Ian Porada

2019-04-25

Learning speed and accuracy are of universal interest for reinforcement learning problems. In this paper, we investigate meta-learning approaches for adaptation of the trace decay parameter {\lambda} used in TD({\lambda}), from the perspective of optimizing a bias-variance tradeoff. We propose an off-policy applicable method of meta-learning the {\lambda} parameters via optimizing a metaobjective with effcient incremental updates. The proposed trust-region style algorithm, under proper assumptions, is shown to be equivalent to optimizing the bias-variance tradeoff for the overall target for all states. In experiments, we validate the effectiveness of the proposed method MTA showing its significantly faster and more accurate learning patterns compared to the compared methods and baselines.

Time Series Simulation by Conditional Generative Adversarial Net (1904.11419v1)

Rao Fu, Jie Chen, Shutian Zeng, Yiping Zhuang, Agus Sudjianto

2019-04-25

Generative Adversarial Net (GAN) has been proven to be a powerful machine learning tool in image data analysis and generation. In this paper, we propose to use Conditional Generative Adversarial Net (CGAN) to learn and simulate time series data. The conditions can be both categorical and continuous variables containing different kinds of auxiliary information. Our simulation studies show that CGAN is able to learn different kinds of normal and heavy tail distributions, as well as dependent structures of different time series and it can further generate conditional predictive distributions consistent with the training data distributions. We also provide an in-depth discussion on the rationale of GAN and the neural network as hierarchical splines to draw a clear connection with the existing statistical method for distribution generation. In practice, CGAN has a wide range of applications in the market risk and counterparty risk analysis: it can be applied to learn the historical data and generate scenarios for the calculation of Value-at-Risk (VaR) and Expected Shortfall (ES) and predict the movement of the market risk factors. We present a real data analysis including a backtesting to demonstrate CGAN is able to outperform the Historic Simulation, a popular method in market risk analysis for the calculation of VaR. CGAN can also be applied in the economic time series modeling and forecasting, and an example of hypothetical shock analysis for economic models and the generation of potential CCAR scenarios by CGAN is given at the end of the paper.

A Bayesian Approach for the Robust Optimisation of Expensive-To-Evaluate Functions (1904.11416v1)

Nicholas D. Sanders, Richard M. Everson, Jonathan E. Fieldsend, Alma A. M. Rahat

2019-04-25

Many expensive black-box optimisation problems are sensitive to their inputs. In these problems it makes more sense to locate a region of good designs, than a single, possible fragile, optimal design. Expensive black-box functions can be optimised effectively with Bayesian optimisation, where a Gaussian process is a popular choice as a prior over the expensive function. We propose a method for robust optimisation using Bayesian optimisation to find a region of design space in which the expensive function's performance is insensitive to the inputs whilst retaining a good quality. This is achieved by sampling realisations from a Gaussian process modelling the expensive function and evaluating the improvement for each realisation. The expectation of these improvements can be optimised cheaply with an evolutionary algorithm to determine the next location at which to evaluate the expensive function. We describe an efficient process to locate the optimum expected improvement. We show empirically that evaluating the expensive function at the location in the candidate sweet spot about which the model is most uncertain or at random yield the best convergence in contrast to exploitative schemes. We illustrate our method on six test functions in two, five, and ten dimensions, and demonstrate that it is able to outperform a state-of-the-art approach from the literature.

---