Search Results for author:
Found 24 papers, 12 papers with code
The Uniqueness of LLaMA3-70B with Per-Channel Quantization: An Empirical Study
We have observed a distinctive quantization-related behavior in the LLaMA3/3. 1-70B models that is absent in both the LLaMA2-70B and LLaMA3/3. 1-8B/405B models.
Data Overfitting for On-Device Super-Resolution with Dynamic Algorithm and Compiler Co-Design
By splitting videos into chunks and applying a super-resolution (SR) model to overfit each chunk, this scheme of SR models plus video chunks is able to replace traditional video transmission to enhance video quality and transmission efficiency.
Towards High-Quality and Efficient Video Super-Resolution via Spatial-Temporal Data Overfitting
To reconcile such, we propose a novel method for high-quality and efficient video resolution upscaling tasks, which leverages the spatial-temporal information to accurately divide video into chunks, thus keeping the number of chunks as well as the model size to minimum.
DISCO: Distributed Inference with Sparse Communications
For example, each layer of a ResNet-50 model can be distributively inferred across two nodes with five times less data communications, almost half overall computations and half memory requirement for a single node, and achieve comparable accuracy to the original ResNet-50 model.
Pruning Parameterization With Bi-Level Optimization for Efficient Semantic Segmentation on the Edge
With the ever-increasing popularity of edge devices, it is necessary to implement real-time segmentation on the edge for autonomous driving and many other applications.
All-in-One: A Highly Representative DNN Pruning Framework for Edge Devices with Dynamic Power Management
By re-configuring the model to the corresponding pruning ratio for a specific execution frequency (and voltage), we are able to achieve stable inference speed, i. e., keeping the difference in speed performance under various execution frequencies as small as possible.
Self-Ensemble Protection: Training Checkpoints Are Good Data Protectors
In this paper, we uncover them by model checkpoints' gradients, forming the proposed self-ensemble protection (SEP), which is very effective because (1) learning on examples ignored during normal training tends to yield DNNs ignoring normal examples; (2) checkpoints' cross-model gradients are close to orthogonal, meaning that they are as diverse as DNNs with different architectures.
Peeling the Onion: Hierarchical Reduction of Data Redundancy for Efficient Vision Transformer Training
Vision transformers (ViTs) have recently obtained success in many applications, but their intensive computation and heavy memory usage at both training and inference time limit their generalization.
Data Level Lottery Ticket Hypothesis for Vision Transformers
That is, there exists a subset of input image patches such that a ViT can be trained from scratch by using only this subset of patches and achieve similar accuracy to the ViTs trained by using all image patches.
Compiler-Aware Neural Architecture Search for On-Mobile Real-time Super-Resolution
Instead of measuring the speed on mobile devices at each iteration during the search process, a speed model incorporated with compiler optimizations is leveraged to predict the inference latency of the SR block with various width configurations for faster convergence.
CHEX: CHannel EXploration for CNN Model Compression
However, conventional pruning methods have limitations in that: they are restricted to pruning process only, and they require a fully pre-trained large model.
SPViT: Enabling Faster Vision Transformers via Soft Token Pruning
Moreover, our framework can guarantee the identified model to meet resource specifications of mobile devices and FPGA, and even achieve the real-time execution of DeiT-T on mobile platforms.
Ranked #4 on
Efficient ViTs
on ImageNet-1K (with DeiT-S)
Compact Multi-level Sparse Neural Networks with Input Independent Dynamic Rerouting
Deep neural networks (DNNs) have shown to provide superb performance in many real life applications, but their large computation cost and storage requirement have prevented them from being deployed to many edge and internet-of-things (IoT) devices.
Load-balanced Gather-scatter Patterns for Sparse Deep Neural Networks
Our experiments show that GS patterns consistently make better trade-offs between accuracy and computation efficiency compared to conventional structured sparse patterns.
MEST: Accurate and Fast Memory-Economic Sparse Training Framework on the Edge
Systematical evaluation on accuracy, training speed, and memory footprint are conducted, where the proposed MEST framework consistently outperforms representative SOTA works.
HFSP: A Hardware-friendly Soft Pruning Framework for Vision Transformers
Recently, Vision Transformer (ViT) has continuously established new milestones in the computer vision field, while the high computation and memory cost makes its propagation in industrial production difficult.
Sanity Checks for Lottery Tickets: Does Your Winning Ticket Really Win the Jackpot?
Based on our analysis, we summarize a guideline for parameter settings in regards of specific architecture characteristics, which we hope to catalyze the research progress on the topic of lottery ticket hypothesis.
Effective Model Sparsification by Scheduled Grow-and-Prune Methods
It addresses the shortcomings of the previous works by repeatedly growing a subset of layers to dense and then pruning them back to sparse after some training.
Computation on Sparse Neural Networks: an Inspiration for Future Hardware
We show that for practically complicated problems, it is more beneficial to search large and sparse models in the weight dominated region.
A Unified DNN Weight Compression Framework Using Reweighted Optimization Methods
To address the large model size and intensive computation requirement of deep neural networks (DNNs), weight pruning techniques have been proposed and generally fall into two categories, i. e., static regularization-based pruning and dynamic regularization-based pruning.
Learning in the Frequency Domain
Experiment results show that learning in the frequency domain with static channel selection can achieve higher accuracy than the conventional spatial downsampling approach and meanwhile further reduce the input data size.
Non-Volatile Memory Array Based Quantization- and Noise-Resilient LSTM Neural Networks
In cloud and edge computing models, it is important that compute devices at the edge be as power efficient as possible.
Noisy Computations during Inference: Harmful or Helpful?
During inference, by injecting proper noise to signals in the neural networks, the robustness of adversarially-trained neural networks against black-box attacks has been further enhanced by 0. 5% and 1. 13% for two adversarially trained models for MNIST and CIFAR10, respectively.
Training Recurrent Neural Networks against Noisy Computations during Inference
One of the motivations for looking into this problem is to reduce the high power cost of conventional computing of neural network operations through the use of analog neuromorphic circuits.
