Search Results for author:
Found 96 papers, 40 papers with code
Do Counterfactual Examples Complicate Adversarial Training?
We leverage diffusion models to study the robustness-performance tradeoff of robust classifiers.
Min-K%++: Improved Baseline for Detecting Pre-Training Data from Large Language Models
The problem of pre-training data detection for large language models (LLMs) has received growing attention due to its implications in critical issues like copyright violation and test data contamination.
Unsolvable Problem Detection: Evaluating Trustworthiness of Vision Language Models
This paper introduces a novel and significant challenge for Vision Language Models (VLMs), termed Unsolvable Problem Detection (UPD).
Vox-Fusion++: Voxel-based Neural Implicit Dense Tracking and Mapping with Multi-maps
In this paper, we introduce Vox-Fusion++, a multi-maps-based robust dense tracking and mapping system that seamlessly fuses neural implicit representations with traditional volumetric fusion techniques.
Peeking Behind the Curtains of Residual Learning
The utilization of residual learning has become widespread in deep and scalable neural nets.
Adversarial Estimation of Topological Dimension with Harmonic Score Maps
Quantification of the number of variables needed to locally explain complex data is often the first step to better understanding it.
SD-NAE: Generating Natural Adversarial Examples with Stable Diffusion
Natural Adversarial Examples (NAEs), images arising naturally from the environment and capable of deceiving classifiers, are instrumental in robustly evaluating and identifying vulnerabilities in trained models.
DistDNAS: Search Efficient Feature Interactions within 2 Hours
To optimize search efficiency, DistDNAS distributes the search and aggregates the choice of optimal interaction modules on varying data dates, achieving over 25x speed-up and reducing search cost from 2 days to 2 hours.
Farthest Greedy Path Sampling for Two-shot Recommender Search
FGPS enhances path diversity to facilitate more comprehensive supernet exploration, while emphasizing path quality to ensure the effective identification and utilization of promising architectures.
Depth Completion with Multiple Balanced Bases and Confidence for Dense Monocular SLAM
In this paper, we propose a novel method that integrates a light-weight depth completion network into a sparse SLAM system using a multi-basis depth representation, so that dense mapping can be performed online even on a mobile phone.
Stable and Causal Inference for Discriminative Self-supervised Deep Visual Representations
In recent years, discriminative self-supervised methods have made significant strides in advancing various visual tasks.
OpenOOD v1.5: Enhanced Benchmark for Out-of-Distribution Detection
Out-of-Distribution (OOD) detection is critical for the reliable operation of open-world intelligent systems.
Out-of-Distribution Detection
Out of Distribution (OOD) Detection
SIO: Synthetic In-Distribution Data Benefits Out-of-Distribution Detection
Building up reliable Out-of-Distribution (OOD) detectors is challenging, often requiring the use of OOD data during training.
MAPSeg: Unified Unsupervised Domain Adaptation for Heterogeneous Medical Image Segmentation Based on 3D Masked Autoencoding and Pseudo-Labeling
In this study, we introduce Masked Autoencoding and Pseudo-Labeling Segmentation (MAPSeg), a $\textbf{unified}$ UDA framework with great versatility and superior performance for heterogeneous and volumetric medical image segmentation.
Disentangling Learning Representations with Density Estimation
Disentangled learning representations have promising utility in many applications, but they currently suffer from serious reliability issues.
GDOD: Effective Gradient Descent using Orthogonal Decomposition for Multi-Task Learning
GDOD decomposes gradients into task-shared and task-conflict components explicitly and adopts a general update rule for avoiding interference across all task gradients.
HCE: Improving Performance and Efficiency with Heterogeneously Compressed Neural Network Ensemble
We propose a new prespective on exploring the intrinsic diversity within a model architecture to build efficient DNN ensemble.
PIDS: Joint Point Interaction-Dimension Search for 3D Point Cloud
In this work, we establish PIDS, a novel paradigm to jointly explore point interactions and point dimensions to serve semantic segmentation on point cloud data.
Ranked #6 on
Robust 3D Semantic Segmentation
on SemanticKITTI-C
Neural Architecture Search
Robust 3D Semantic Segmentation
+1
More Generalized and Personalized Unsupervised Representation Learning In A Distributed System
In this work, we propose a novel method, FedStyle, to learn a more generalized global model by infusing local style information with local content information for contrastive learning, and to learn more personalized local models by inducing local style information for downstream tasks.
Preserving background sound in noise-robust voice conversion via multi-task learning
Background sound is an informative form of art that is helpful in providing a more immersive experience in real-application voice conversion (VC) scenarios.
Vox-Fusion: Dense Tracking and Mapping with Voxel-based Neural Implicit Representation
In this work, we present a dense tracking and mapping system named Vox-Fusion, which seamlessly fuses neural implicit representations with traditional volumetric fusion methods.
Approximate Computing and the Efficient Machine Learning Expedition
In this work, we enlighten the synergistic nature of AxC and ML and elucidate the impact of AxC in designing efficient ML systems.
Fed-CBS: A Heterogeneity-Aware Client Sampling Mechanism for Federated Learning via Class-Imbalance Reduction
Based on this measure, we also design a computation-efficient client sampling strategy, such that the actively selected clients will generate a more class-balanced grouped dataset with theoretical guarantees.
NashAE: Disentangling Representations through Adversarial Covariance Minimization
We present a self-supervised method to disentangle factors of variation in high-dimensional data that does not rely on prior knowledge of the underlying variation profile (e. g., no assumptions on the number or distribution of the individual latent variables to be extracted).
Fine-grain Inference on Out-of-Distribution Data with Hierarchical Classification
Furthermore, we diagnose the classifiers performance at each level of the hierarchy improving the explainability and interpretability of the models predictions.
FADE: Enabling Federated Adversarial Training on Heterogeneous Resource-Constrained Edge Devices
However, the high demand for memory capacity and computing power makes large-scale federated adversarial training infeasible on resource-constrained edge devices.
Learning and Compositionality: a Unification Attempt via Connectionist Probabilistic Programming
We consider learning and compositionality as the key mechanisms towards simulating human-like intelligence.
Tunable Hybrid Proposal Networks for the Open World
Current state-of-the-art object proposal networks are trained with a closed-world assumption, meaning they learn to only detect objects of the training classes.
Vox-Surf: Voxel-based Implicit Surface Representation
Virtual content creation and interaction play an important role in modern 3D applications such as AR and VR.
NASRec: Weight Sharing Neural Architecture Search for Recommender Systems
To overcome the data multi-modality and architecture heterogeneity challenges in the recommendation domain, NASRec establishes a large supernet (i. e., search space) to search the full architectures.
Privacy Leakage of Adversarial Training Models in Federated Learning Systems
Adversarial Training (AT) is crucial for obtaining deep neural networks that are robust to adversarial attacks, yet recent works found that it could also make models more vulnerable to privacy attacks.
IQDUBBING: Prosody modeling based on discrete self-supervised speech representation for expressive voice conversion
Specifically, prosody vector is first extracted from pre-trained VQ-Wav2Vec model, where rich prosody information is embedded while most speaker and environment information are removed effectively by quantization.
FL-WBC: Enhancing Robustness against Model Poisoning Attacks in Federated Learning from a Client Perspective
Furthermore, we derive a certified robustness guarantee against model poisoning attacks and a convergence guarantee to FedAvg after applying our FL-WBC.
A Layer-wise Adversarial-aware Quantization Optimization for Improving Robustness
After quantization, the cost can be greatly saved, and the quantized models are more hardware friendly with acceptable accuracy loss.
Global Vision Transformer Pruning with Hessian-Aware Saliency
This work aims on challenging the common design philosophy of the Vision Transformer (ViT) model with uniform dimension across all the stacked blocks in a model stage, where we redistribute the parameters both across transformer blocks and between different structures within the block via the first systematic attempt on global structural pruning.
Automated Mobile Attention KPConv Networks via A Wide & Deep Predictor
MAKPConv employs a depthwise kernel to reduce resource consumption and re-calibrates the contribution of kernel points towards each neighbor point via Neighbor-Kernel attention to improve representation power.
Privacy-Preserving Representation Learning on Graphs: A Mutual Information Perspective
Existing representation learning methods on graphs have achieved state-of-the-art performance on various graph-related tasks such as node classification, link prediction, etc.
Soteria: Provable Defense Against Privacy Leakage in Federated Learning From Representation Perspective
The key idea of our defense is learning to perturb data representation such that the quality of the reconstructed data is severely degraded, while FL performance is maintained.
Enriching Source Style Transfer in Recognition-Synthesis based Non-Parallel Voice Conversion
Specifically, prosodic features are used to explicit model prosody, while VAE and reference encoder are used to implicitly model prosody, which take Mel spectrum and bottleneck feature as input respectively.
Mixture Outlier Exposure: Towards Out-of-Distribution Detection in Fine-grained Environments
We then propose Mixture Outlier Exposure (MixOE), which mixes ID data and training outliers to expand the coverage of different OOD granularities, and trains the model such that the prediction confidence linearly decays as the input transitions from ID to OOD.
Medical Image Classification
Out-of-Distribution Detection
+1
Neuromorphic Algorithm-hardware Codesign for Temporal Pattern Learning
This circuit implementation of the neuron model is simulated to demonstrate its ability to react to temporal spiking patterns with an adaptive threshold.
The Multi-speaker Multi-style Voice Cloning Challenge 2021
The challenge consists of two tracks, namely few-shot track and one-shot track, where the participants are required to clone multiple target voices with 100 and 5 samples respectively.
FedCor: Correlation-Based Active Client Selection Strategy for Heterogeneous Federated Learning
In this work, we propose FedCor -- an FL framework built on a correlation-based client selection strategy, to boost the convergence rate of FL.
Can Targeted Adversarial Examples Transfer When the Source and Target Models Have No Label Space Overlap?
During the online phase of the attack, we then leverage representations of highly related proxy classes from the whitebox distribution to fool the blackbox model into predicting the desired target class.
The Untapped Potential of Off-the-Shelf Convolutional Neural Networks
Over recent years, a myriad of novel convolutional network architectures have been developed to advance state-of-the-art performance on challenging recognition tasks.
A Case for 3D Integrated System Design for Neuromorphic Computing & AI Applications
Over the last decade, artificial intelligence has found many applications areas in the society.
BSQ: Exploring Bit-Level Sparsity for Mixed-Precision Neural Network Quantization
Mixed-precision quantization can potentially achieve the optimal tradeoff between performance and compression rate of deep neural networks, and thus, have been widely investigated.
On Provable Backdoor Defense in Collaborative Learning
The framework shows that the subset selection process, a deciding factor for subset aggregation methods, can be viewed as a code design problem.
Improving Adversarial Robustness in Weight-quantized Neural Networks
Neural networks are getting deeper and more computation-intensive nowadays.
GraphFL: A Federated Learning Framework for Semi-Supervised Node Classification on Graphs
However, existing FL methods 1) perform poorly when data across clients are non-IID, 2) cannot handle data with new label domains, and 3) cannot leverage unlabeled data, while all these issues naturally happen in real-world graph-based problems.
Provable Defense against Privacy Leakage in Federated Learning from Representation Perspective
In this work, we show our key observation that the data representation leakage from gradients is the essential cause of privacy leakage in FL.
ScaleNAS: One-Shot Learning of Scale-Aware Representations for Visual Recognition
We use ScaleNAS to create high-resolution models for two different tasks, ScaleNet-P for human pose estimation and ScaleNet-S for semantic segmentation.
Ranked #5 on
Multi-Person Pose Estimation
on COCO test-dev
Fast IR Drop Estimation with Machine Learning
IR drop constraint is a fundamental requirement enforced in almost all chip designs.
DVERGE: Diversifying Vulnerabilities for Enhanced Robust Generation of Ensembles
The process is hard, often requires models with large capacity, and suffers from significant loss on clean data accuracy.
Efficient, Direct, and Restricted Black-Box Graph Evasion Attacks to Any-Layer Graph Neural Networks via Influence Function
Specifically, we first introduce two influence functions, i. e., feature-label influence and label influence, that are defined on GNNs and label propagation (LP), respectively.
Reinforcement Learning-based Black-Box Evasion Attacks to Link Prediction in Dynamic Graphs
Link prediction in dynamic graphs (LPDG) is an important research problem that has diverse applications such as online recommendations, studies on disease contagion, organizational studies, etc.
LotteryFL: Personalized and Communication-Efficient Federated Learning with Lottery Ticket Hypothesis on Non-IID Datasets
Rather than learning a shared global model in classic federated learning, each client learns a personalized model via LotteryFL; the communication cost can be significantly reduced due to the compact size of lottery networks.
NASGEM: Neural Architecture Search via Graph Embedding Method
To preserve graph correlation information in encoding, we propose NASGEM which stands for Neural Architecture Search via Graph Embedding Method.
Defending against GAN-based Deepfake Attacks via Transformation-aware Adversarial Faces
On the one hand, the quality of the synthesized faces is reduced with more visual artifacts such that the synthesized faces are more obviously fake or less convincing to human observers.
PENNI: Pruned Kernel Sharing for Efficient CNN Inference
Although state-of-the-art (SOTA) CNNs achieve outstanding performance on various tasks, their high computation demand and massive number of parameters make it difficult to deploy these SOTA CNNs onto resource-constrained devices.
Learning Low-rank Deep Neural Networks via Singular Vector Orthogonality Regularization and Singular Value Sparsification
In this work, we propose SVD training, the first method to explicitly achieve low-rank DNNs during training without applying SVD on every step.
Neural Predictor for Neural Architecture Search
First we train N random architectures to generate N (architecture, validation accuracy) pairs and use them to train a regression model that predicts accuracy based on the architecture.
AutoShrink: A Topology-aware NAS for Discovering Efficient Neural Architecture
Specifically, both ShrinkCNN and ShrinkRNN are crafted within 1. 5 GPU hours, which is 7. 2x and 6. 7x faster than the crafting time of SOTA CNN and RNN models, respectively.
Structural sparsification for Far-field Speaker Recognition with GNA
Recently, deep neural networks (DNN) have been widely used in speaker recognition area.
Defending Neural Backdoors via Generative Distribution Modeling
An original trigger used by an attacker to build the backdoored model represents only a point in the space.
Conditional Transferring Features: Scaling GANs to Thousands of Classes with 30% Less High-quality Data for Training
As such, training our GAN architecture requires much fewer high-quality images with a small number of additional low-quality images.
Exploring Bit-Slice Sparsity in Deep Neural Networks for Efficient ReRAM-Based Deployment
However, the power hungry analog-to-digital converters (ADCs) prevent the practical deployment of ReRAM-based DNN accelerators on end devices with limited chip area and power budget.
Towards Efficient and Secure Delivery of Data for Deep Learning with Privacy-Preserving
When using MoLe for VGG-16 network on CIFAR dataset, the computational overhead is only 9% and the data transmission overhead is 5. 12%.
DASNet: Dynamic Activation Sparsity for Neural Network Efficiency Improvement
In addition to conventional compression techniques, e. g., weight pruning and quantization, removing unimportant activations can reduce the amount of data communication and the computation cost.
Feedback Learning for Improving the Robustness of Neural Networks
Recent research studies revealed that neural networks are vulnerable to adversarial attacks.
DeepHoyer: Learning Sparser Neural Network with Differentiable Scale-Invariant Sparsity Measures
Inspired by the Hoyer measure (the ratio between L1 and L2 norms) used in traditional compressed sensing problems, we present DeepHoyer, a set of sparsity-inducing regularizers that are both differentiable almost everywhere and scale-invariant.
Joint Regularization on Activations and Weights for Efficient Neural Network Pruning
With the rapid scaling up of deep neural networks (DNNs), extensive research studies on network model compression such as weight pruning have been performed for improving deployment efficiency.
SwiftNet: Using Graph Propagation as Meta-knowledge to Search Highly Representative Neural Architectures
Designing neural architectures for edge devices is subject to constraints of accuracy, inference latency, and computational cost.
Snooping Attacks on Deep Reinforcement Learning
In these snooping threat models, the adversary does not have the ability to interact with the target agent's environment, and can only eavesdrop on the action and reward signals being exchanged between agent and environment.
Integral Pruning on Activations and Weights for Efficient Neural Networks
With the rapidly scaling up of deep neural networks (DNNs), extensive research studies on network model compression such as weight pruning have been performed for efficient deployment.
HyPar: Towards Hybrid Parallelism for Deep Learning Accelerator Array
In this paper, inspired by recent work in machine learning systems, we propose a solution HyPar to determine layer-wise parallelism for deep neural network training with an array of DNN accelerators.
Towards Leveraging the Information of Gradients in Optimization-based Adversarial Attack
We intuitively and empirically prove the rationality of our method in reducing the search space.
Adversarial Attacks for Optical Flow-Based Action Recognition Classifiers
The success of deep learning research has catapulted deep models into production systems that our society is becoming increasingly dependent on, especially in the image and video domains.
LEASGD: an Efficient and Privacy-Preserving Decentralized Algorithm for Distributed Learning
Distributed learning systems have enabled training large-scale models over large amount of data in significantly shorter time.
Differentiable Fine-grained Quantization for Deep Neural Network Compression
Thus judiciously selecting different precision for different layers/structures can potentially produce more efficient models compared to traditional quantization methods by striking a better balance between accuracy and compression rate.
Towards Efficient and Secure Delivery of Data for Training and Inference with Privacy-Preserving
When using MoLe for VGG-16 network on CIFAR dataset, the computational overhead is only 9% and the data transmission overhead is 5. 12%.
DPatch: An Adversarial Patch Attack on Object Detectors
Successful realization of DPatch also illustrates the intrinsic vulnerability of the modern detector architectures to such patch-based adversarial attacks.
SmoothOut: Smoothing Out Sharp Minima to Improve Generalization in Deep Learning
It becomes an open question whether escaping sharp minima can improve the generalization.
Learning Intrinsic Sparse Structures within Long Short-Term Memory
This work aims to learn structurally-sparse Long Short-Term Memory (LSTM) by reducing the sizes of basic structures within LSTM units, including input updates, gates, hidden states, cell states and outputs.
GraphR: Accelerating Graph Processing Using ReRAM
GRAPHR gains a speedup of 1. 16x to 4. 12x, and is 3. 67x to 10. 96x more energy efficiency compared to PIM-based architecture.
Distributed, Parallel, and Cluster Computing Hardware Architecture
MAT: A Multi-strength Adversarial Training Method to Mitigate Adversarial Attacks
Our experiments show that different adversarial strengths, i. e., perturbation levels of adversarial examples, have different working zones to resist the attack.
TernGrad: Ternary Gradients to Reduce Communication in Distributed Deep Learning
We mathematically prove the convergence of TernGrad under the assumption of a bound on gradients.
Coordinating Filters for Faster Deep Neural Networks
Moreover, Force Regularization better initializes the low-rank DNNs such that the fine-tuning can converge faster toward higher accuracy.
A Compact DNN: Approaching GoogLeNet-Level Accuracy of Classification and Domain Adaptation
Our DNN has 4. 1M parameters, which is only 6. 7% of AlexNet or 59% of GoogLeNet.
Generative Poisoning Attack Method Against Neural Networks
A countermeasure is also designed to detect such poisoning attack methods by checking the loss of the target model.
Group Scissor: Scaling Neuromorphic Computing Design to Large Neural Networks
Following rank clipping, group connection deletion further reduces the routing area of LeNet and ConvNet to 8. 1\% and 52. 06\%, respectively.
Classification Accuracy Improvement for Neuromorphic Computing Systems with One-level Precision Synapses
Brain inspired neuromorphic computing has demonstrated remarkable advantages over traditional von Neumann architecture for its high energy efficiency and parallel data processing.
Learning Structured Sparsity in Deep Neural Networks
SSL can: (1) learn a compact structure from a bigger DNN to reduce computation cost; (2) obtain a hardware-friendly structured sparsity of DNN to efficiently accelerate the DNNs evaluation.
Faster CNNs with Direct Sparse Convolutions and Guided Pruning
Pruning CNNs in a way that increase inference speed often imposes specific sparsity structures, thus limiting the achievable sparsity levels.
A New Learning Method for Inference Accuracy, Core Occupation, and Performance Co-optimization on TrueNorth Chip
IBM TrueNorth chip uses digital spikes to perform neuromorphic computing and achieves ultrahigh execution parallelism and power efficiency.
