United States

Pruning has been an effective solution to reduce the number of computations and the memory requirement in deep learning.
The pruning unit plays an important role in exploiting the GPU resources efficiently. 
The filter is proposed as a simple pruning unit of structured pruning.
However, since the filter is quite large as pruning unit, the accuracy drop is considerable with a high pruning ratio.
GPU rearranges the weight and input tensors into tiles (blocks) for efficient computation. 
To fully utilize GPU resources, this tile structure should be considered, which is the goal of block pruning. 
However, previous block pruning prunes both row vectors and column vectors. 
Pruning of row vectors in a tile corresponds to filter pruning, and it also interferes with column-wise block pruning of the following layer.
In contrast, column vectors are much smaller than row vectors and can achieve lower accuracy drop.
Additionally, if the pruning ratio for each tile is different,
GPU utilization can be limited by imbalanced workloads by irregular-sized blocks.
The same pruning ratio for the weight tiles processed in parallel enables the actual inference process to fully utilize the resources without idle time.
This paper proposes balanced column-wise block pruning, named BCBP, to satisfy two conditions: the column-wise minimal size of the pruning unit and balanced workloads. 
We demonstrate that BCBP is superior to previous pruning methods through comprehensive experiments.

AAAI 2023

Balanced Column-Wise Block Pruning for Maximizing GPU Parallelism

workload balancing

tiling

block pruning

The purpose of the AAAI conference series is to promote research in Artificial Intelligence (AI) and foster scientific exchange between researchers, practitioners, scientists, students, and engineers across the entirety of AI and its affiliated disciplines. AAAI-23 is the Thirty-Seventh AAAI Conference on Artificial Intelligence. The theme of this conference is to create collaborative bridges within and beyond AI. Like previous AAAI conferences, AAAI-23 will feature technical paper presentations, special tracks, invited speakers, workshops, tutorials, poster sessions, senior member presentations, competitions, and exhibit programs, and two new activities: a Bridge Program and a Lab Program. Many of these activities are tailored to the theme of bridges and all are selected according to the highest standards, with additional programs for students and young researchers. 
AAAI is providing you with a conference planner, which you can use to help organize your itinerary of activities. This includes talks to attend in person, talks to attend remotely, breaks with colleagues and your site seeing activities. To access this conference planner, please go to [https://aaai-2023.takemobi.io](https://aaai-2023.takemobi.io).

In order to access this site, you need to register. If you haven't already, please register [here](https://aaai.org/Conferences/AAAI-23/registration/).


The purpose of the AAAI conference series is to promote research in Artificial Intelligence (AI) and foster scientific exchange between researchers, practitioners, scientists, students, and engineers across the entirety of AI and its affiliated disciplines.

poster

Weakly supervised detection of anomalies in surveillance videos is a challenging task. Going beyond existing works that have deficient capabilities to localize anomalies in long videos, we propose a novel glance and focus network to effectively integrate spatial-temporal information for accurate anomaly detection. In addition, we empirically found that existing approaches that use feature magnitudes to represent the degree of anomalies typically ignore the effects of scene variations, and hence result in sub-optimal performance due to the inconsistency of feature magnitudes across scenes. To address this issue, we propose the Feature Amplification Mechanism and a Magnitude Contrastive Loss to enhance the discriminativeness of feature magnitudes for detecting anomalies. Experimental results on two large-scale benchmarks UCF-Crime and XD-Violence manifest that our method outperforms state-of-the-art approaches.

MGFN : Magnitude-Contrastive Glance-and-Focus Network for Weakly-Supervised Video Anomaly Detection

The creation of a parameterized stylized character involves careful selection of numerous parameters, also known as the "avatar vectors" that can be interpreted by the avatar engine. Existing unsupervised avatar vector estimation methods that auto-create avatars for users, however, often fail to work because of the domain gap between realistic faces and stylized avatar images. To this end, we propose SwiftAvatar, a novel avatar auto-creation framework that is evidently superior to previous works. SwiftAvatar introduces dual-domain generators to create pairs of realistic faces and avatar images using shared latent codes. The latent codes can then be bridged with the avatar vectors as pairs, by performing GAN inversion on the avatar images rendered from the engine using avatar vectors. Through this way, we are able to synthesize paired data in high-quality as many as possible, consisting of avatar vectors and their corresponding realistic faces. We also propose semantic augmentation to improve the diversity of synthesis. Finally, a light-weight avatar vector estimator is trained on the synthetic pairs to implement efficient auto-creation. Our experiments demonstrate the effectiveness and efficiency of SwiftAvatar on two different avatar engines. The superiority and advantageous flexibility of SwiftAvatar are also verified in both subjective and objective evaluations.

SwiftAvatar: Efficient Auto-Creation of Parameterized Stylized Character on Arbitrary Avatar Engines

Gaze estimator computes the gaze direction based on face images. Most existing gaze estimation methods perform well under within-dataset settings, but can not generalize to unseen domains. In particular, the ground-truth labels in unseen domain are often unavailable. In this paper, we propose a new domain generalization method based on gaze-consistent features. Our idea is to consider the gaze-irrelevant factors as unfavorable interference and disturb the training data against them, so that the model cannot fit to these gaze-irrelevant factors, instead, only fits to the gaze-consistent features. To this end, we first disturb the training data via adversarial attack or data augmentation based on the gaze-irrelevant factors, i.e., identity, expression, illumination and tone. Then we extract the gaze-consistent features by aligning the gaze features from disturbed data with non-disturbed gaze features. Experimental results show that our proposed method achieves state-of-the-art performance on gaze domain generalization task. Furthermore, our proposed method also improves domain adaption performance on gaze estimation. Our work provides new insight on gaze domain generalization task.

Learning a Generalized Gaze Estimator from Gaze-Consistent Feature

In the scene of self-supervised graph learning, Mutual Information (MI) was recently introduced for graph encoding to generate robust node embeddings. A successful representative is Deep Graph Infomax (DGI), which essentially operates on the space of node features but ignores topological structures, and just considers global graph summary. In this paper, we present an effective model called Deep Graph Structural Infomax (DGSI) to learn node representation. We explore to derive the structural mutual information from the perspective of Information Bottleneck (IB), which defines a trade-off between the sufficiency and minimality of representation on the condition of the topological structure preservation. Intuitively, the derived constraints formally maximize the structural mutual information both edge-wise and local neighborhood-wise. Besides, we develop a general framework that incorporates the global representational mutual information, local representational mutual information, and sufficient structural information into the node representation. Essentially, our DGSI extends DGI and could capture more fine-grained semantic information as well as beneficial structural information in a self-supervised manner, thereby improving node representation and further boosting the learning performance. Extensive experiments on different types of datasets demonstrate the effectiveness and superiority of the proposed method.

Deep Graph Structural Infomax

3D automatic annotation has received increased attention since manually annotating 3D point clouds is laborious. However, existing methods are usually complicated, e.g., pipelined training for 3D foreground/background segmentation, cylindrical object proposals and point completion. Furthermore, they often overlook the inter-object feature correlation that is particularly informative to hard samples for 3D annotation. 
To this end, we propose a simple yet effective end-to-end Context-Aware Transformer (CAT) as an automated 3D-box labeler to generate precise 3D box annotations from 2D boxes, trained with a small number of human annotations. We adopt the general encoder-decoder architecture, where the CAT encoder consists of an intra-object encoder (local) and an inter-object encoder (global), performing self-attention along the sequence and batch dimensions, respectively. The former models intra-object interactions among points and the latter extracts feature relations among different objects, thus boosting scene-level understanding.
Via local and global encoders, CAT can generate high-quality 3D box annotations with a streamlined workflow, allowing it to outperform existing state-of-the-arts by up to 1.79% 3D AP on the hard task of the KITTI test set.

Context-Aware Transformer for 3D Point Cloud Automatic Annotation

Self-attention-based networks have obtained impressive performance in parallel training and global context modeling. However, it is weak in local dependency capturing, especially for data with strong local correlations such as utterances. Therefore, we will mine linguistic information of the original text based on a semantic dependency and the semantic relationship between nodes is regarded as prior knowledge to revise the distribution of self-attention. On the other hand, given the strong correlation between input characters, we introduce a one-dimensional (1-D) convolution neural network (CNN) producing query(Q) and value(V) in the self-attention mechanism for a better fusion of local contextual information. Then, we migrate this variant of the self-attention networks to speech synthesis tasks and propose a non-autoregressive (NAR) neural Text-to-Speech (TTS): SeDepTTS. Experimental results show that our model yields good performance in speech synthesis. Specifically, the proposed method yields significant improvement for the processing of pause, stress, and intonation in speech.

SeDepTTSÔºö Enhancing the Naturalness via Semantic Dependency and Local Convolution for Text-to-Speech Synthesis

Cell clustering is a critical step in analyzing single-cell RNA sequencing (scRNA-seq) data, which allows us to characterize the cellular heterogeneity of transcriptional profiling at the single-cell level. Single-cell deep embedded representation models have recently become popular since they can learn feature representation and clustering simultaneously. However, the model still suffers from a variety of significant challenges, including the massive amount of data, pervasive dropout events, and complicated noise patterns in transcriptional profiling. Here, we propose a Single-Cell Deep Embedding Fusion Representation (scDEFR) model, which develop a deep embedded fusion representation to learn fused heterogeneous latent embedding that contains both the transcriptome gene-level information and the cell topology information. We first fuse them layer by layer to obtain compressed representations of intercellular relationships and transcriptome information. After that, the zero-inflated negative binomial model (ZINB)-based decoder is proposed to capture the global probabilistic structure of the data and reconstruct the final gene expression information and cell graph. Finally, by simultaneously integrating the clustering loss, crossentropy loss, ZINB loss, and the cell graph reconstruction loss,
scDEFR can optimize clustering performance and learn the latent representation in fused information under a joint mutual supervised strategy. We conducted extensive and comprehensive experiments on 15 single-cell RNA-seq datasets from different sequencing platforms to demonstrate the superiority of scDEFR over a variety of state-of-the-art methods. 


Unsupervised Deep Embedded Fusion Representation of Single-cell Transcriptomics

Cross-view geo-localization aims to estimate the location of a query ground image by matching it to a reference geo-tagged aerial images database. As an extremely challenging task, its difficulties root in the drastic view changes and different capturing time between two views. Despite these difficulties, recent works achieve outstanding progress on cross-view geo-localization benchmarks. However, existing methods still suffer from poor performance on the cross-area benchmarks, in which the training and testing data are captured from two different regions. We attribute this deficiency to the lack of ability to extract the spatial configuration of visual feature layouts and models' overfitting on low-level details from the training set. In this paper, we propose GeoDTR which explicitly disentangles geometric information from raw features and learns the spatial correlations among visual features from aerial and ground pairs with a novel geometric layout extractor module. This module generates a set of geometric layout descriptors, modulating the raw features and producing high-quality latent representations. In addition, we elaborate on two categories of data augmentations, (i) Layout simulation, which varies the spatial configuration while keeping the low-level details intact. (ii) Semantic augmentation, which alters the low-level details and encourages the model to capture spatial configurations. These augmentations help to improve the performance of the cross-view geo-localization models, especially on the cross-area benchmarks. Moreover, we propose a counterfactual-based learning process to benefit the geometric layout extractor in exploring spatial information. Extensive experiments show that GeoDTR not only achieves state-of-the-art results but also significantly boosts the performance on same-area and cross-area benchmarks. Our code can be found at https://gitlab.com/vail-uvm/geodtr.

Cross-view Geo-localization via Learning Disentangled Geometric Layout Correspondence

Federated learning aims to collaboratively train models without accessing their client's local private data. The data may be Non-IID for different clients and thus resulting in poor performance. Recently, personalized federated learning (PFL) has achieved great success in handling Non-IID data by enforcing regularization in local optimization or improving the model aggregation scheme on the server. However, most of the PFL approaches do not take into account the unfair competition issue caused by the imbalanced data distribution and lack of positive samples for some classes in each client. To address this issue, we propose a novel and generic PFL framework termed Federated Averaging via Binary Classification, dubbed FedABC. In particular, we adopt the ``one-vs-all'' training strategy in each client to alleviate the unfair competition between classes by constructing a personalized binary classification problem for each class. This may aggravate the class imbalance challenge and thus a novel personalized binary classification loss that incorporates both the under-sampling and hard sample mining strategies is designed. Extensive experiments are conducted on two popular datasets under different settings, and the results demonstrate that our FedABC can significantly outperform the existing counterparts.

FedABC: Targeting Fair Competition in Personalized Federated Learning

Access to a representative sample from the population is an assumption that underpins all of machine learning. Selection effects can cause observations to instead come from a subpopulation, by which our inferences may be subject to bias. 
It is therefore important to know whether or not a sample is affected by selection effects. We study under which conditions we can identify selection bias and give results for both parametric and non-parametric families of distributions. Based on these results we develop two practical methods to determine whether or not an observed sample comes from a distribution subject to selection bias. Through extensive evaluation on synthetic and real world data we verify that our methods beat the state of the art both in detecting as well as characterizing selection bias.

Balanced Column-Wise Block Pruning for Maximizing GPU Parallelism

Downloads

Next from AAAI 2023

MGFN : Magnitude-Contrastive Glance-and-Focus Network for Weakly-Supervised Video Anomaly Detection

Similar lecture

Balanced Column-Wise Block Pruning for Maximizing GPU Parallelism