Abstract
To classify in-distribution samples, deep neural networks learn label-discriminative representations, which, however, are not necessarily distribution-discriminative according to the information bottleneck. Therefore, trained networks could assign unexpected high-confidence predictions to out-of-distribution samples drawn from distributions differing from that of in-distribution samples. Specifically, networks extract the strongly label-related information from in-distribution samples to learn the label-discriminative representations but discard the weakly label-related information. Accordingly, networks treat out-of-distribution samples with minimum label-sensitive information as in-distribution samples. According to the different informativeness properties of in- and out-of-distribution samples, a Dual Representation Learning (DRL) method learns distribution-discriminative representations that are weakly related to the labeling of in-distribution samples and combines label- and distribution-discriminative representations to detect out-of-distribution samples. For a label-discriminative representation, DRL constructs the complementary distribution-discriminative representation by an implicit constraint, i.e., integrating diverse intermediate representations where an intermediate representation less similar to the label-discriminative representation owns a higher weight. Experiments show that DRL outperforms the state-of-the-art methods for out-of-distribution detection.
Breaking Down Out-of-Distribution Detection: Many Methods Based on OOD Training Data Estimate a Combination of the Same Core Quantities
Authors: Julian Bitterwolf, Alexander Meinke, Maximilian Augustin, Matthias Hein
Subjects: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
Abstract
It is an important problem in trustworthy machine learning to recognize out-of-distribution (OOD) inputs which are inputs unrelated to the in-distribution task. Many out-of-distribution detection methods have been suggested in recent years. The goal of this paper is to recognize common objectives as well as to identify the implicit scoring functions of different OOD detection methods. We focus on the sub-class of methods that use surrogate OOD data during training in order to learn an OOD detection score that generalizes to new unseen out-distributions at test time. We show that binary discrimination between in- and (different) out-distributions is equivalent to several distinct formulations of the OOD detection problem. When trained in a shared fashion with a standard classifier, this binary discriminator reaches an OOD detection performance similar to that of Outlier Exposure. Moreover, we show that the confidence loss which is used by Outlier Exposure has an implicit scoring function which differs in a non-trivial fashion from the theoretically optimal scoring function in the case where training and test out-distribution are the same, which again is similar to the one used when training an Energy-Based OOD detector or when adding a background class. In practice, when trained in exactly the same way, all these methods perform similarly.
Keyword: expected calibration error
There is no result
Keyword: overconfident
There is no result
Keyword: overconfidence
There is no result
Keyword: confidence
Context-aware Proposal Network for Temporal Action Detection
Abstract
This technical report presents our first place winning solution for temporal action detection task in CVPR-2022 AcitivityNet Challenge. The task aims to localize temporal boundaries of action instances with specific classes in long untrimmed videos. Recent mainstream attempts are based on dense boundary matchings and enumerate all possible combinations to produce proposals. We argue that the generated proposals contain rich contextual information, which may benefits detection confidence prediction. To this end, our method mainly consists of the following three steps: 1) action classification and feature extraction by Slowfast, CSN, TimeSformer, TSP, I3D-flow, VGGish-audio, TPN and ViViT; 2) proposal generation. Our proposed Context-aware Proposal Network (CPN) builds on top of BMN, GTAD and PRN to aggregate contextual information by randomly masking some proposal features. 3) action detection. The final detection prediction is calculated by assigning the proposals with corresponding video-level classifcation results. Finally, we ensemble the results under different feature combination settings and achieve 45.8% performance on the test set, which improves the champion result in CVPR-2021 ActivityNet Challenge by 1.1% in terms of average mAP.
Redundancy Reduction Twins Network: A Training framework for Multi-output Emotion Regression
Authors: Xin Jing, Meishu Song, Andreas Triantafyllopoulos, Zijiang Yang, Björn W. Schuller
Subjects: Sound (cs.SD); Audio and Speech Processing (eess.AS)
Abstract
In this paper, we propose the Redundancy Reduction Twins Network (RRTN), a redundancy reduction training framework that minimizes redundancy by measuring the cross-correlation matrix between the outputs of the same network fed with distorted versions of a sample and bringing it as close to the identity matrix as possible. RRTN also applies a new loss function, the Barlow Twins loss function, to help maximize the similarity of representations obtained from different distorted versions of a sample. However, as the distribution of losses can cause performance fluctuations in the network, we also propose the use of a Restrained Uncertainty Weight Loss (RUWL) or joint training to identify the best weights for the loss function. Our best approach on CNN14 with the proposed methodology obtains a CCC over emotion regression of 0.678 on the ExVo Multi-task dev set, a 4.8% increase over a vanilla CNN 14 CCC of 0.647, which achieves a significant difference at the 95% confidence interval (2-tailed).
Thompson Sampling for (Combinatorial) Pure Exploration
Abstract
Existing methods of combinatorial pure exploration mainly focus on the UCB approach. To make the algorithm efficient, they usually use the sum of upper confidence bounds within arm set $S$ to represent the upper confidence bound of $S$, which can be much larger than the tight upper confidence bound of $S$ and leads to a much higher complexity than necessary, since the empirical means of different arms in $S$ are independent. To deal with this challenge, we explore the idea of Thompson Sampling (TS) that uses independent random samples instead of the upper confidence bounds, and design the first TS-based algorithm TS-Explore for (combinatorial) pure exploration. In TS-Explore, the sum of independent random samples within arm set $S$ will not exceed the tight upper confidence bound of $S$ with high probability. Hence it solves the above challenge, and achieves a lower complexity upper bound than existing efficient UCB-based algorithms in general combinatorial pure exploration. As for pure exploration of classic multi-armed bandit, we show that TS-Explore achieves an asymptotically optimal complexity upper bound.
Gray Learning from Non-IID Data with Out-of-distribution Samples
Authors: Zhilin Zhao, Longbing Cao, Chang-Dong Wang
Abstract
The quality of the training data annotated by experts cannot be guaranteed, even more so for non-IID data consisting of both in- and out-of-distribution samples (i.e., in-distribution and out-of-distribution samples hold different distributions). Experts may mistakenly annotate out-of-distribution samples the same as in-distribution samples, incurring untrustworthy ground-truth labels. Learning such non-IID data mixing in- and out-of-distribution samples with untrustworthy labels significantly challenges both shallow and deep learning, with no relevant work reported. It would be possible to identify trustworthy complementary labels of a sample indicating which classes it does not belong to, because both in- and out-of-distribution samples do not belong to the classes except those corresponding to the ground-truth label. With this insight, we propose a novel \textit{gray learning} approach to robustly learn from non-IID data with both in- and out-of-distribution samples. Due to the uncertain distributions of training samples, we reject the complementary labels for low-confidence inputs while mapping high-confidence inputs to the ground-truth labels in training. Building on the statistical learning theory, we derive the generalization error which shows that gray learning achieves a tight bound on the non-IID data. Extensive experiments show that our method provides significant improvement over alternative methods from robust statistics.
Supervision Adaptation Balances In-Distribution Generalization and Out-of-Distribution Detection
Abstract
When there is a discrepancy between in-distribution (ID) samples and out-of-distribution (OOD) samples, deep neural networks trained on ID samples suffer from high-confidence prediction on OOD samples. This is primarily caused by unavailable OOD samples to constrain the networks in the training process. To improve the OOD sensitivity of deep networks, several state-of-the-art methods introduce samples from other real-world datasets as OOD samples to the training process and assign manually-determined labels to these OOD samples. However, they sacrifice the classification accuracy because the unreliable labeling of OOD samples would disrupt ID classification. To balance ID generalization and OOD detection, a major challenge to tackle is to make OOD samples compatible with ID ones, which is addressed by our proposed \textit{supervision adaptation} method in this paper to define adaptive supervision information for OOD samples. First, by measuring the dependency between ID samples and their labels through mutual information, we reveal the form of the supervision information in terms of the negative probabilities of all classes. Second, after exploring the data correlations between ID and OOD samples by solving multiple binary regression problems, we estimate the supervision information to make ID classes more separable. We perform experiments on four advanced network architectures with two ID datasets and eleven OOD datasets to demonstrate the balancing effect of our supervision adaptation method in achieving both the ID classification ability and the OOD detection capacity.
Out-of-distribution Detection by Cross-class Vicinity Distribution of In-distribution Data
Abstract
Deep neural networks only learn to map in-distribution inputs to their corresponding ground truth labels in the training phase without differentiating out-of-distribution samples from in-distribution ones. This results from the assumption that all samples are independent and identically distributed without distributional distinction. Therefore, a pretrained network learned from the in-distribution samples treats out-of-distribution samples as in-distribution and makes high-confidence predictions on them in the test phase. To address this issue, we draw out-of-distribution samples from the vicinity distribution of training in-distribution samples for learning to reject the prediction on out-of-distribution inputs. A \textit{Cross-class Vicinity Distribution} is introduced by assuming that an out-of-distribution sample generated by mixing multiple in-distribution samples does not share the same classes of its constituents. We thus improve the discriminability of a pretrained network by finetuning it with out-of-distribution samples drawn from the cross-class vicinity distribution, where each out-of-distribution input corresponds to a complementary label. Experiments on various in-/out-of-distribution datasets show that the proposed method significantly outperforms existing methods in improving the capacity of discriminating between in- and out-of-distribution samples.
Label and Distribution-discriminative Dual Representation Learning for Out-of-Distribution Detection
Abstract
To classify in-distribution samples, deep neural networks learn label-discriminative representations, which, however, are not necessarily distribution-discriminative according to the information bottleneck. Therefore, trained networks could assign unexpected high-confidence predictions to out-of-distribution samples drawn from distributions differing from that of in-distribution samples. Specifically, networks extract the strongly label-related information from in-distribution samples to learn the label-discriminative representations but discard the weakly label-related information. Accordingly, networks treat out-of-distribution samples with minimum label-sensitive information as in-distribution samples. According to the different informativeness properties of in- and out-of-distribution samples, a Dual Representation Learning (DRL) method learns distribution-discriminative representations that are weakly related to the labeling of in-distribution samples and combines label- and distribution-discriminative representations to detect out-of-distribution samples. For a label-discriminative representation, DRL constructs the complementary distribution-discriminative representation by an implicit constraint, i.e., integrating diverse intermediate representations where an intermediate representation less similar to the label-discriminative representation owns a higher weight. Experiments show that DRL outperforms the state-of-the-art methods for out-of-distribution detection.
Data-Driven Synthesis of Symbolic Abstractions with Guaranteed Confidence
Abstract
In this work, we propose a data-driven approach for the construction of finite abstractions (a.k.a., symbolic models) for discrete-time deterministic control systems with unknown dynamics. We leverage notions of so-called alternating bisimulation functions (ABF), as a relation between each unknown system and its symbolic model, to quantify the mismatch between state behaviors of two systems. Accordingly, one can employ our proposed results to perform formal verification and synthesis over symbolic models and then carry the results back over unknown original systems. In our data-driven setting, we first cast the required conditions for constructing ABF as a robust optimization program (ROP). Solving the provided ROP is not tractable due to the existence of unknown models in the constraints of ROP. To tackle this difficulty, we collect finite numbers of data from trajectories of unknown systems and propose a scenario optimization program (SOP) corresponding to the original ROP. By establishing a probabilistic relation between optimal values of SOP and ROP, we formally construct ABF between unknown systems and their symbolic models based on the number of data and a required confidence level. We verify the effectiveness of our data-driven results over two physical case studies with unknown models including (i) a DC motor and (ii) a nonlinear jet engine compressor. We construct symbolic models from data as appropriate substitutes of original systems and synthesize policies maintaining states of unknown systems in a safe set within infinite time horizons with some guaranteed confidence levels.
Guardian Angel: A Novel Walking Aid for the Visually Impaired
Authors: Ko-Wei Tai, HuaYen Lee, Hsin-Huei Chen, Jeng-Sheng Yeh, Ming Ouhyoung
Subjects: Human-Computer Interaction (cs.HC); Computer Vision and Pattern Recognition (cs.CV)
Abstract
This work introduces Guardian Angel, an Android App that assists visually impaired people to avoid danger in complex traffic environment. The system, consisting of object detection by pretrained YOLO model, distance estimation and moving direction estimation, provides information about surrounding vehicles and alarms users of potential danger without expensive special purpose device. With an experiment of 8 subjects, we corroborate that in terms of satisfaction score in pedestrian-crossing experiment with the assistance of our App using a smartphone is better than when without under 99% confidence level. The time needed to cross a road is shorter on average with the assistance of our system, however, not reaching significant difference by our experiment. The App has been released in Google Play Store, open to the public for free.
EAGER: Asking and Answering Questions for Automatic Reward Shaping in Language-guided RL
Authors: Thomas Carta, Sylvain Lamprier, Pierre-Yves Oudeyer, Olivier Sigaud
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
Abstract
Reinforcement learning (RL) in long horizon and sparse reward tasks is notoriously difficult and requires a lot of training steps. A standard solution to speed up the process is to leverage additional reward signals, shaping it to better guide the learning process. In the context of language-conditioned RL, the abstraction and generalisation properties of the language input provide opportunities for more efficient ways of shaping the reward. In this paper, we leverage this idea and propose an automated reward shaping method where the agent extracts auxiliary objectives from the general language goal. These auxiliary objectives use a question generation (QG) and question answering (QA) system: they consist of questions leading the agent to try to reconstruct partial information about the global goal using its own trajectory. When it succeeds, it receives an intrinsic reward proportional to its confidence in its answer. This incentivizes the agent to generate trajectories which unambiguously explain various aspects of the general language goal. Our experimental study shows that this approach, which does not require engineer intervention to design the auxiliary objectives, improves sample efficiency by effectively directing exploration.
Breaking Down Out-of-Distribution Detection: Many Methods Based on OOD Training Data Estimate a Combination of the Same Core Quantities
Authors: Julian Bitterwolf, Alexander Meinke, Maximilian Augustin, Matthias Hein
Subjects: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
Abstract
It is an important problem in trustworthy machine learning to recognize out-of-distribution (OOD) inputs which are inputs unrelated to the in-distribution task. Many out-of-distribution detection methods have been suggested in recent years. The goal of this paper is to recognize common objectives as well as to identify the implicit scoring functions of different OOD detection methods. We focus on the sub-class of methods that use surrogate OOD data during training in order to learn an OOD detection score that generalizes to new unseen out-distributions at test time. We show that binary discrimination between in- and (different) out-distributions is equivalent to several distinct formulations of the OOD detection problem. When trained in a shared fashion with a standard classifier, this binary discriminator reaches an OOD detection performance similar to that of Outlier Exposure. Moreover, we show that the confidence loss which is used by Outlier Exposure has an implicit scoring function which differs in a non-trivial fashion from the theoretically optimal scoring function in the case where training and test out-distribution are the same, which again is similar to the one used when training an Energy-Based OOD detector or when adding a background class. In practice, when trained in exactly the same way, all these methods perform similarly.
Test Time Transform Prediction for Open Set Histopathological Image Recognition
Authors: Adrian Galdran. Katherine J. Hewitt, Narmin L. Ghaffari, Jakob N. Kather, Gustavo Carneiro, Miguel A. González Ballester
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Tissue typology annotation in Whole Slide histological images is a complex and tedious, yet necessary task for the development of computational pathology models. We propose to address this problem by applying Open Set Recognition techniques to the task of jointly classifying tissue that belongs to a set of annotated classes, e.g. clinically relevant tissue categories, while rejecting in test time Open Set samples, i.e. images that belong to categories not present in the training set. To this end, we introduce a new approach for Open Set histopathological image recognition based on training a model to accurately identify image categories and simultaneously predict which data augmentation transform has been applied. In test time, we measure model confidence in predicting this transform, which we expect to be lower for images in the Open Set. We carry out comprehensive experiments in the context of colorectal cancer assessment from histological images, which provide evidence on the strengths of our approach to automatically identify samples from unknown categories. Code is released at https://github.com/agaldran/t3po .
Improving Localization for Semi-Supervised Object Detection
Authors: Leonardo Rossi, Akbar Karimi, Andrea Prati
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Nowadays, Semi-Supervised Object Detection (SSOD) is a hot topic, since, while it is rather easy to collect images for creating a new dataset, labeling them is still an expensive and time-consuming task. One of the successful methods to take advantage of raw images on a Semi-Supervised Learning (SSL) setting is the Mean Teacher technique, where the operations of pseudo-labeling by the Teacher and the Knowledge Transfer from the Student to the Teacher take place simultaneously. However, the pseudo-labeling by thresholding is not the best solution since the confidence value is not strictly related to the prediction uncertainty, not permitting to safely filter predictions. In this paper, we introduce an additional classification task for bounding box localization to improve the filtering of the predicted bounding boxes and obtain higher quality on Student training. Furthermore, we empirically prove that bounding box regression on the unsupervised part can equally contribute to the training as much as category classification. Our experiments show that our IL-net (Improving Localization net) increases SSOD performance by 1.14% AP on COCO dataset in limited-annotation regime. The code is available at https://github.com/IMPLabUniPr/unbiased-teacher/tree/ilnet
Enhancing Multi-view Stereo with Contrastive Matching and Weighted Focal Loss
Abstract
Learning-based multi-view stereo (MVS) methods have made impressive progress and surpassed traditional methods in recent years. However, their accuracy and completeness are still struggling. In this paper, we propose a new method to enhance the performance of existing networks inspired by contrastive learning and feature matching. First, we propose a Contrast Matching Loss (CML), which treats the correct matching points in depth-dimension as positive sample and other points as negative samples, and computes the contrastive loss based on the similarity of features. We further propose a Weighted Focal Loss (WFL) for better classification capability, which weakens the contribution of low-confidence pixels in unimportant areas to the loss according to predicted confidence. Extensive experiments performed on DTU, Tanks and Temples and BlendedMVS datasets show our method achieves state-of-the-art performance and significant improvement over baseline network.
Keyword: scaling
MME-CRS: Multi-Metric Evaluation Based on Correlation Re-Scaling for Evaluating Open-Domain Dialogue
Authors: Pengfei Zhang, Xiaohui Hu, Kaidong Yu, Jian Wang, Song Han, Cao Liu, Chunyang Yuan
Abstract
Automatic open-domain dialogue evaluation is a crucial component of dialogue systems. Recently, learning-based evaluation metrics have achieved state-of-the-art performance in open-domain dialogue evaluation. However, these metrics, which only focus on a few qualities, are hard to evaluate dialogue comprehensively. Furthermore, these metrics lack an effective score composition approach for diverse evaluation qualities. To address the above problems, we propose a Multi-Metric Evaluation based on Correlation Re-Scaling (MME-CRS) for evaluating open-domain dialogue. Firstly, we build an evaluation metric composed of 5 groups of parallel sub-metrics called Multi-Metric Evaluation (MME) to evaluate the quality of dialogue comprehensively. Furthermore, we propose a novel score composition method called Correlation Re-Scaling (CRS) to model the relationship between sub-metrics and diverse qualities. Our approach MME-CRS ranks first on the final test data of DSTC10 track5 subtask1 Automatic Open-domain Dialogue Evaluation Challenge with a large margin, which proved the effectiveness of our proposed approach.
Terrain Classification using Transfer Learning on Hyperspectral Images: A Comparative study
Abstract
A Hyperspectral image contains much more number of channels as compared to a RGB image, hence containing more information about entities within the image. The convolutional neural network (CNN) and the Multi-Layer Perceptron (MLP) have been proven to be an effective method of image classification. However, they suffer from the issues of long training time and requirement of large amounts of the labeled data, to achieve the expected outcome. These issues become more complex while dealing with hyperspectral images. To decrease the training time and reduce the dependence on large labeled dataset, we propose using the method of transfer learning. The hyperspectral dataset is preprocessed to a lower dimension using PCA, then deep learning models are applied to it for the purpose of classification. The features learned by this model are then used by the transfer learning model to solve a new classification problem on an unseen dataset. A detailed comparison of CNN and multiple MLP architectural models is performed, to determine an optimum architecture that suits best the objective. The results show that the scaling of layers not always leads to increase in accuracy but often leads to overfitting, and also an increase in the training time.The training time is reduced to greater extent by applying the transfer learning approach rather than just approaching the problem by directly training a new model on large datasets, without much affecting the accuracy.
Agricultural Plantation Classification using Transfer Learning Approach based on CNN
Authors: Uphar Singh, Tushar Musale, Ranjana Vyas, O.P.Vyas (Indian Institute of Information Technology, Allahabad, India)
Abstract
Hyper-spectral images are images captured from a satellite that gives spatial and spectral information of specific region.A Hyper-spectral image contains much more number of channels as compared to a RGB image, hence containing more information about entities within the image. It makes them well suited for the classification of objects in a snap. In the past years, the efficiency of hyper-spectral image recognition has increased significantly with deep learning. The Convolution Neural Network(CNN) and Multi-Layer Perceptron(MLP) has demonstrated to be an excellent process of classifying images. However, they suffer from the issues of long training time and requirement of large amounts of the labeled data, to achieve the expected outcome. These issues become more complex while dealing with hyper-spectral images. To decrease the training time and reduce the dependence on large labeled data-set, we propose using the method of transfer learning.The features learned by CNN and MLP models are then used by the transfer learning model to solve a new classification problem on an unseen dataset. A detailed comparison of CNN and multiple MLP architectural models is performed, to determine an optimum architecture that suits best the objective. The results show that the scaling of layers not always leads to increase in accuracy but often leads to over-fitting, and also an increase in the training time.The training time is reduced to greater extent by applying the transfer learning approach rather than just approaching the problem by directly training a new model on large data-sets, without much affecting the accuracy.
Phoebe: QoS-Aware Distributed Stream Processing through Anticipating Dynamic Workloads
Authors: Morgan K. Geldenhuys, Dominik Scheinert, Odej Kao, Lauritz Thamsen
Subjects: Distributed, Parallel, and Cluster Computing (cs.DC)
Abstract
Distributed Stream Processing systems have become an essential part of big data processing platforms. They are characterized by the high-throughput processing of near to real-time event streams with the goal of delivering low-latency results and thus enabling time-sensitive decision making. At the same time, results are expected to be consistent even in the presence of partial failures where exactly-once processing guarantees are required for correctness. Stream processing workloads are oftentimes dynamic in nature which makes static configurations highly inefficient as time goes by. Static resource allocations will almost certainly either negatively impact upon the Quality of Service and/or result in higher operational costs. In this paper we present Phoebe, a proactive approach to system auto-tuning for Distributed Stream Processing jobs executing on dynamic workloads. Our approach makes use of parallel profiling runs, QoS modeling, and runtime optimization to provide a general solution whereby configuration parameters are automatically tuned to ensure a stable service as well as alignment with recovery time Quality of Service targets. Phoebe makes use of Time Series Forecasting to gain an insight into future workload requirements thereby delivering scaling decisions which are accurate, long-lived, and reliable. Our experiments demonstrate that Phoebe is able to deliver a stable service while at the same time reducing resource over-provisioning.
Understanding RowHammer Under Reduced Wordline Voltage: An Experimental Study Using Real DRAM Devices
Authors: A. Giray Yağlıkçı, Haocong Luo, Geraldo F. de Oliviera, Ataberk Olgun, Minesh Patel, Jisung Park, Hasan Hassan, Jeremie S. Kim, Lois Orosa, Onur Mutlu
Subjects: Hardware Architecture (cs.AR); Cryptography and Security (cs.CR)
Abstract
RowHammer is a circuit-level DRAM vulnerability, where repeatedly activating and precharging a DRAM row, and thus alternating the voltage of a row's wordline between low and high voltage levels, can cause bit flips in physically nearby rows. Recent DRAM chips are more vulnerable to RowHammer: with technology node scaling, the minimum number of activate-precharge cycles to induce a RowHammer bit flip reduces and the RowHammer bit error rate increases. Therefore, it is critical to develop effective and scalable approaches to protect modern DRAM systems against RowHammer. To enable such solutions, it is essential to develop a deeper understanding of the RowHammer vulnerability of modern DRAM chips. However, even though the voltage toggling on a wordline is a key determinant of RowHammer vulnerability, no prior work experimentally demonstrates the effect of wordline voltage (VPP) on the RowHammer vulnerability. Our work closes this gap in understanding. This is the first work to experimentally demonstrate on 272 real DRAM chips that lowering VPP reduces a DRAM chip's RowHammer vulnerability. We show that lowering VPP 1) increases the number of activate-precharge cycles needed to induce a RowHammer bit flip by up to 85.8% with an average of 7.4% across all tested chips and 2) decreases the RowHammer bit error rate by up to 66.9% with an average of 15.2% across all tested chips. At the same time, reducing VPP marginally worsens a DRAM cell's access latency, charge restoration, and data retention time within the guardbands of system-level nominal timing parameters for 208 out of 272 tested chips. We conclude that reducing VPP is a promising strategy for reducing a DRAM chip's RowHammer vulnerability without requiring modifications to DRAM chips.
Limitations of the NTK for Understanding Generalization in Deep Learning
Abstract
The ``Neural Tangent Kernel'' (NTK) (Jacot et al 2018), and its empirical variants have been proposed as a proxy to capture certain behaviors of real neural networks. In this work, we study NTKs through the lens of scaling laws, and demonstrate that they fall short of explaining important aspects of neural network generalization. In particular, we demonstrate realistic settings where finite-width neural networks have significantly better data scaling exponents as compared to their corresponding empirical and infinite NTKs at initialization. This reveals a more fundamental difference between the real networks and NTKs, beyond just a few percentage points of test accuracy. Further, we show that even if the empirical NTK is allowed to be pre-trained on a constant number of samples, the kernel scaling does not catch up to the neural network scaling. Finally, we show that the empirical NTK continues to evolve throughout most of the training, in contrast with prior work which suggests that it stabilizes after a few epochs of training. Altogether, our work establishes concrete limitations of the NTK approach in understanding generalization of real networks on natural datasets.
QuAFL: Federated Averaging Can Be Both Asynchronous and Communication-Efficient
Authors: Hossein Zakerinia, Shayan Talaei, Giorgi Nadiradze, Dan Alistarh
Abstract
Federated Learning (FL) is an emerging paradigm to enable the large-scale distributed training of machine learning models, while still providing privacy guarantees. In this work, we jointly address two of the main practical challenges when scaling federated optimization to large node counts: the need for tight synchronization between the central authority and individual computing nodes, and the large communication cost of transmissions between the central server and clients. Specifically, we present a new variant of the classic federated averaging (FedAvg) algorithm, which supports both asynchronous communication and communication compression. We provide a new analysis technique showing that, in spite of these system relaxations, our algorithm essentially matches the best known bounds for FedAvg, under reasonable parameter settings. On the experimental side, we show that our algorithm ensures fast practical convergence for standard federated tasks.
Sensitivity of Average Precision to Bounding Box Perturbations
Authors: Ali Borji
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Object detection is a fundamental vision task. It has been highly researched in academia and has been widely adopted in industry. Average Precision (AP) is the standard score for evaluating object detectors. Our understanding of the subtleties of this score, however, is limited. Here, we quantify the sensitivity of AP to bounding box perturbations and show that AP is very sensitive to small translations. Only one pixel shift is enough to drop the mAP of a model by 8.4%. The mAP drop over small objects with only one pixel shift is 23.1%. The corresponding numbers when ground-truth (GT) boxes are used as predictions are 23% and 41.7%, respectively. These results explain why achieving higher mAP becomes increasingly harder as models get better. We also investigate the effect of box scaling on AP. Code and data is available at https://github.com/aliborji/AP_Box_Perturbation.
EpiGRAF: Rethinking training of 3D GANs
Authors: Ivan Skorokhodov, Sergey Tulyakov, Yiqun Wang, Peter Wonka
Abstract
A very recent trend in generative modeling is building 3D-aware generators from 2D image collections. To induce the 3D bias, such models typically rely on volumetric rendering, which is expensive to employ at high resolutions. During the past months, there appeared more than 10 works that address this scaling issue by training a separate 2D decoder to upsample a low-resolution image (or a feature tensor) produced from a pure 3D generator. But this solution comes at a cost: not only does it break multi-view consistency (i.e. shape and texture change when the camera moves), but it also learns the geometry in a low fidelity. In this work, we show that it is possible to obtain a high-resolution 3D generator with SotA image quality by following a completely different route of simply training the model patch-wise. We revisit and improve this optimization scheme in two ways. First, we design a location- and scale-aware discriminator to work on patches of different proportions and spatial positions. Second, we modify the patch sampling strategy based on an annealed beta distribution to stabilize training and accelerate the convergence. The resulted model, named EpiGRAF, is an efficient, high-resolution, pure 3D generator, and we test it on four datasets (two introduced in this work) at $256^2$ and $512^2$ resolutions. It obtains state-of-the-art image quality, high-fidelity geometry and trains ${\approx} 2.5 \times$ faster than the upsampler-based counterparts. Project website: https://universome.github.io/epigraf.
Vicinity Vision Transformer
Authors: Weixuan Sun, Zhen Qin, Hui Deng, Jianyuan Wang, Yi Zhang, Kaihao Zhang, Nick Barnes, Stan Birchfield, Lingpeng Kong, Yiran Zhong
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Vision transformers have shown great success on numerous computer vision tasks. However, its central component, softmax attention, prohibits vision transformers from scaling up to high-resolution images, due to both the computational complexity and memory footprint being quadratic. Although linear attention was introduced in natural language processing (NLP) tasks to mitigate a similar issue, directly applying existing linear attention to vision transformers may not lead to satisfactory results. We investigate this problem and find that computer vision tasks focus more on local information compared with NLP tasks. Based on this observation, we present a Vicinity Attention that introduces a locality bias to vision transformers with linear complexity. Specifically, for each image patch, we adjust its attention weight based on its 2D Manhattan distance measured by its neighbouring patches. In this case, the neighbouring patches will receive stronger attention than far-away patches. Moreover, since our Vicinity Attention requires the token length to be much larger than the feature dimension to show its efficiency advantages, we further propose a new Vicinity Vision Transformer (VVT) structure to reduce the feature dimension without degenerating the accuracy. We perform extensive experiments on the CIFAR100, ImageNet1K, and ADE20K datasets to validate the effectiveness of our method. Our method has a slower growth rate of GFlops than previous transformer-based and convolution-based networks when the input resolution increases. In particular, our approach achieves state-of-the-art image classification accuracy with 50% fewer parameters than previous methods.
Controllability of Coarsely Measured Networked Linear Dynamical Systems (Extended Version)
Authors: Nafiseh Ghoroghchian, Rajasekhar Anguluri, Gautam Dasarathy, Stark C. Draper
Subjects: Systems and Control (eess.SY); Machine Learning (cs.LG); Signal Processing (eess.SP); Machine Learning (stat.ML)
Abstract
We consider the controllability of large-scale linear networked dynamical systems when complete knowledge of network structure is unavailable and knowledge is limited to coarse summaries. We provide conditions under which average controllability of the fine-scale system can be well approximated by average controllability of the (synthesized, reduced-order) coarse-scale system. To this end, we require knowledge of some inherent parametric structure of the fine-scale network that makes this type of approximation possible. Therefore, we assume that the underlying fine-scale network is generated by the stochastic block model (SBM) -- often studied in community detection. We then provide an algorithm that directly estimates the average controllability of the fine-scale system using a coarse summary of SBM. Our analysis indicates the necessity of underlying structure (e.g., in-built communities) to be able to quantify accurately the controllability from coarsely characterized networked dynamics. We also compare our method to that of the reduced-order method and highlight the regimes where both can outperform each other. Finally, we provide simulations to confirm our theoretical results for different scalings of network size and density, and the parameter that captures how much community-structure is retained in the coarse summary.
Keyword: calibration
A Surgical Platform for Intracerebral Hemorrhage Robotic Evacuation (ASPIHRE): A Non-metallic MR-guided Concentric Tube Robot
Authors: Anthony L. Gunderman, Saikat Sengupta, Eleni Siampli, Dimitri Sigounas, Christopher Kellner, Chima Oluigbo, Karun Sharma, Isuru Godage, Kevin Cleary, Yue Chen
Abstract
Intracerebral hemorrhage (ICH) is the deadliest stroke sub-type, with a one-month mortality rate as high as 52%. Due to the potential cortical disruption caused by craniotomy, conservative management (watchful waiting) has historically been a common method of treatment. Minimally invasive evacuation has recently become an accepted method of treatment for patients with deep-seated hematoma 30-50 mL in volume, but proper visualization and tool dexterity remain constrained in conventional endoscopic approaches, particularly with larger hematoma volumes (> 50 mL). In this article we describe the development of ASPIHRE (A Surgical Platform for Intracerebral Hemorrhage Robotic Evacuation), the first-ever concentric tube robot that uses off-the-shelf plastic tubes for MR-guided ICH evacuation, improving tool dexterity and procedural visualization. The robot kinematics model is developed based on a calibration-based method and tube mechanics modeling, allowing the models to consider both variable curvature and torsional deflection. The MR-safe pneumatic motors are controlled using a variable gain PID algorithm producing a rotational accuracy of 0.317 +/- 0.3 degrees. The hardware and theoretical models are validated in a series of systematic bench-top and MRI experiments resulting in positional accuracy of the tube tip of 1.39 +- 0.54 mm. Following validation of targeting accuracy, the evacuation efficacy of the robot was tested in an MR-guided phantom clot evacuation experiment. The robot was able to evacuate an initially 38.36 mL clot in 5 minutes, leaving a residual hematoma of 8.14 mL, well below the 15 mL guideline suggesting good post-ICH evacuation clinical outcomes.
Bi-Calibration Networks for Weakly-Supervised Video Representation Learning
Authors: Fuchen Long, Ting Yao, Zhaofan Qiu, Xinmei Tian, Jiebo Luo, Tao Mei
Subjects: Computer Vision and Pattern Recognition (cs.CV); Multimedia (cs.MM)
Abstract
The leverage of large volumes of web videos paired with the searched queries or surrounding texts (e.g., title) offers an economic and extensible alternative to supervised video representation learning. Nevertheless, modeling such weakly visual-textual connection is not trivial due to query polysemy (i.e., many possible meanings for a query) and text isomorphism (i.e., same syntactic structure of different text). In this paper, we introduce a new design of mutual calibration between query and text to boost weakly-supervised video representation learning. Specifically, we present Bi-Calibration Networks (BCN) that novelly couples two calibrations to learn the amendment from text to query and vice versa. Technically, BCN executes clustering on all the titles of the videos searched by an identical query and takes the centroid of each cluster as a text prototype. The query vocabulary is built directly on query words. The video-to-text/video-to-query projections over text prototypes/query vocabulary then start the text-to-query or query-to-text calibration to estimate the amendment to query or text. We also devise a selection scheme to balance the two corrections. Two large-scale web video datasets paired with query and title for each video are newly collected for weakly-supervised video representation learning, which are named as YOVO-3M and YOVO-10M, respectively. The video features of BCN learnt on 3M web videos obtain superior results under linear model protocol on downstream tasks. More remarkably, BCN trained on the larger set of 10M web videos with further fine-tuning leads to 1.6%, and 1.8% gains in top-1 accuracy on Kinetics-400, and Something-Something V2 datasets over the state-of-the-art TDN, and ACTION-Net methods with ImageNet pre-training. Source code and datasets are available at \url{https://github.com/FuchenUSTC/BCN}.
Keyword: out of distribution detection
There is no result
Keyword: out-of-distribution detection
Label and Distribution-discriminative Dual Representation Learning for Out-of-Distribution Detection
Breaking Down Out-of-Distribution Detection: Many Methods Based on OOD Training Data Estimate a Combination of the Same Core Quantities
Keyword: expected calibration error
There is no result
Keyword: overconfident
There is no result
Keyword: overconfidence
There is no result
Keyword: confidence
Context-aware Proposal Network for Temporal Action Detection
Redundancy Reduction Twins Network: A Training framework for Multi-output Emotion Regression
Thompson Sampling for (Combinatorial) Pure Exploration
Gray Learning from Non-IID Data with Out-of-distribution Samples
Supervision Adaptation Balances In-Distribution Generalization and Out-of-Distribution Detection
Out-of-distribution Detection by Cross-class Vicinity Distribution of In-distribution Data
Label and Distribution-discriminative Dual Representation Learning for Out-of-Distribution Detection
Data-Driven Synthesis of Symbolic Abstractions with Guaranteed Confidence
Guardian Angel: A Novel Walking Aid for the Visually Impaired
EAGER: Asking and Answering Questions for Automatic Reward Shaping in Language-guided RL
Breaking Down Out-of-Distribution Detection: Many Methods Based on OOD Training Data Estimate a Combination of the Same Core Quantities
Test Time Transform Prediction for Open Set Histopathological Image Recognition
Improving Localization for Semi-Supervised Object Detection
Enhancing Multi-view Stereo with Contrastive Matching and Weighted Focal Loss
Keyword: scaling
MME-CRS: Multi-Metric Evaluation Based on Correlation Re-Scaling for Evaluating Open-Domain Dialogue
Terrain Classification using Transfer Learning on Hyperspectral Images: A Comparative study
Agricultural Plantation Classification using Transfer Learning Approach based on CNN
Phoebe: QoS-Aware Distributed Stream Processing through Anticipating Dynamic Workloads
Understanding RowHammer Under Reduced Wordline Voltage: An Experimental Study Using Real DRAM Devices
Limitations of the NTK for Understanding Generalization in Deep Learning
QuAFL: Federated Averaging Can Be Both Asynchronous and Communication-Efficient
Sensitivity of Average Precision to Bounding Box Perturbations
EpiGRAF: Rethinking training of 3D GANs
Vicinity Vision Transformer
Controllability of Coarsely Measured Networked Linear Dynamical Systems (Extended Version)
Keyword: calibration
A Surgical Platform for Intracerebral Hemorrhage Robotic Evacuation (ASPIHRE): A Non-metallic MR-guided Concentric Tube Robot
Bi-Calibration Networks for Weakly-Supervised Video Representation Learning