no code implementations • 1 Apr 2024 • Han Cai, Muyang Li, Zhuoyang Zhang, Qinsheng Zhang, Ming-Yu Liu, Song Han
In parallel to prior conditional control methods, CAN controls the image generation process by dynamically manipulating the weight of the neural network.
1 code implementation • 29 Feb 2024 • Muyang Li, Tianle Cai, Jiaxin Cao, Qinsheng Zhang, Han Cai, Junjie Bai, Yangqing Jia, Ming-Yu Liu, Kai Li, Song Han
To overcome this dilemma, we observe the high similarity between the input from adjacent diffusion steps and propose displaced patch parallelism, which takes advantage of the sequential nature of the diffusion process by reusing the pre-computed feature maps from the previous timestep to provide context for the current step.
1 code implementation • 7 Feb 2024 • Zhuoyang Zhang, Han Cai, Song Han
For the training, we begin with the knowledge distillation from the SAM-ViT-H image encoder to EfficientViT.
no code implementations • ICCV 2023 • Han Cai, Junyan Li, Muyan Hu, Chuang Gan, Song Han
Without performance loss on Cityscapes, our EfficientViT provides up to 8. 8x and 3. 8x GPU latency reduction over SegFormer and SegNeXt, respectively.
no code implementations • 25 Sep 2022 • Jeremie Houssineau, Han Cai, Murat Uney, Emmanuel Delande
Fusing and sharing information from multiple sensors over a network is a challenging task.
5 code implementations • 29 May 2022 • Han Cai, Junyan Li, Muyan Hu, Chuang Gan, Song Han
Without performance loss on Cityscapes, our EfficientViT provides up to 13. 9$\times$ and 6. 2$\times$ GPU latency reduction over SegFormer and SegNeXt, respectively.
Ranked #24 on Semantic Segmentation on Cityscapes val
1 code implementation • CVPR 2022 • Yihan Wang, Muyang Li, Han Cai, Wei-Ming Chen, Song Han
Inspired by this finding, we design LitePose, an efficient single-branch architecture for pose estimation, and introduce two simple approaches to enhance the capacity of LitePose, including Fusion Deconv Head and Large Kernel Convs.
Ranked #5 on Multi-Person Pose Estimation on MS COCO (Validation AP metric)
no code implementations • 25 Apr 2022 • Han Cai, Ji Lin, Yujun Lin, Zhijian Liu, Haotian Tang, Hanrui Wang, Ligeng Zhu, Song Han
Deep neural networks (DNNs) have achieved unprecedented success in the field of artificial intelligence (AI), including computer vision, natural language processing and speech recognition.
no code implementations • NeurIPS 2021 • Ji Lin, Wei-Ming Chen, Han Cai, Chuang Gan, Song Han
We further propose receptive field redistribution to shift the receptive field and FLOPs to the later stage and reduce the computation overhead.
1 code implementation • 28 Oct 2021 • Ji Lin, Wei-Ming Chen, Han Cai, Chuang Gan, Song Han
We further propose network redistribution to shift the receptive field and FLOPs to the later stage and reduce the computation overhead.
no code implementations • ICLR 2022 • Han Cai, Chuang Gan, Ji Lin, Song Han
We introduce Network Augmentation (NetAug), a new training method for improving the performance of tiny neural networks.
1 code implementation • NeurIPS 2020 • Han Cai, Chuang Gan, Ligeng Zhu, Song Han
Furthermore, combined with feature extractor adaptation, TinyTL provides 7. 3-12. 9x memory saving without sacrificing accuracy compared to fine-tuning the full Inception-V3.
1 code implementation • CVPR 2020 • Tianzhe Wang, Kuan Wang, Han Cai, Ji Lin, Zhijian Liu, Song Han
However, training this quantization-aware accuracy predictor requires collecting a large number of quantized <model, accuracy> pairs, which involves quantization-aware finetuning and thus is highly time-consuming.
4 code implementations • ACL 2020 • Hanrui Wang, Zhanghao Wu, Zhijian Liu, Han Cai, Ligeng Zhu, Chuang Gan, Song Han
To enable low-latency inference on resource-constrained hardware platforms, we propose to design Hardware-Aware Transformers (HAT) with neural architecture search.
Ranked #21 on Machine Translation on WMT2014 English-French
1 code implementation • ICLR 2020 • Han Cai, Chuang Gan, Tianzhe Wang, Zhekai Zhang, Song Han
Most of the traditional approaches either manually design or use neural architecture search (NAS) to find a specialized neural network and train it from scratch for each case, which is computationally expensive and unscalable.
10 code implementations • 26 Aug 2019 • Han Cai, Chuang Gan, Tianzhe Wang, Zhekai Zhang, Song Han
On diverse edge devices, OFA consistently outperforms state-of-the-art (SOTA) NAS methods (up to 4. 0% ImageNet top1 accuracy improvement over MobileNetV3, or same accuracy but 1. 5x faster than MobileNetV3, 2. 6x faster than EfficientNet w. r. t measured latency) while reducing many orders of magnitude GPU hours and $CO_2$ emission.
Ranked #76 on Neural Architecture Search on ImageNet
no code implementations • 24 Apr 2019 • Song Han, Han Cai, Ligeng Zhu, Ji Lin, Kuan Wang, Zhijian Liu, Yujun Lin
Moreover, we shorten the design cycle by 200x than previous work, so that we can afford to design specialized neural network models for different hardware platforms.
23 code implementations • ICLR 2019 • Han Cai, Ligeng Zhu, Song Han
We address the high memory consumption issue of differentiable NAS and reduce the computational cost (GPU hours and GPU memory) to the same level of regular training while still allowing a large candidate set.
Ranked #6 on Neural Architecture Search on CIFAR-10 Image Classification (using extra training data)
no code implementations • 14 Nov 2018 • Haifeng Zhang, Zilong Guo, Han Cai, Chris Wang, Wei-Nan Zhang, Yong Yu, Wenxin Li, Jun Wang
With the rapid growth of the express industry, intelligent warehouses that employ autonomous robots for carrying parcels have been widely used to handle the vast express volume.
no code implementations • 14 Nov 2018 • Haokun Chen, Xinyi Dai, Han Cai, Wei-Nan Zhang, Xuejian Wang, Ruiming Tang, Yuzhou Zhang, Yong Yu
Reinforcement learning (RL) has recently been introduced to interactive recommender systems (IRS) because of its nature of learning from dynamic interactions and planning for long-run performance.
3 code implementations • ICML 2018 • Han Cai, Jiacheng Yang, Wei-Nan Zhang, Song Han, Yong Yu
We introduce a new function-preserving transformation for efficient neural architecture search.
3 code implementations • 2 Dec 2017 • Lianmin Zheng, Jiacheng Yang, Han Cai, Wei-Nan Zhang, Jun Wang, Yong Yu
Unlike previous research platforms on single or multi-agent reinforcement learning, MAgent focuses on supporting the tasks and the applications that require hundreds to millions of agents.
Multi-agent Reinforcement Learning reinforcement-learning +1
6 code implementations • 24 Sep 2017 • Jiaxian Guo, Sidi Lu, Han Cai, Wei-Nan Zhang, Yong Yu, Jun Wang
Automatically generating coherent and semantically meaningful text has many applications in machine translation, dialogue systems, image captioning, etc.
Ranked #1 on Text Generation on COCO Captions
3 code implementations • 16 Jul 2017 • Han Cai, Tianyao Chen, Wei-Nan Zhang, Yong Yu, Jun Wang
Techniques for automatically designing deep neural network architectures such as reinforcement learning based approaches have recently shown promising results.
Ranked #140 on Image Classification on CIFAR-10
2 code implementations • ICLR 2018 • Zhiming Zhou, Han Cai, Shu Rong, Yuxuan Song, Kan Ren, Wei-Nan Zhang, Yong Yu, Jun Wang
Our proposed model also outperforms the baseline methods in the new metric.
1 code implementation • 10 Jan 2017 • Han Cai, Kan Ren, Wei-Nan Zhang, Kleanthis Malialis, Jun Wang, Yong Yu, Defeng Guo
In this paper, we formulate the bid decision process as a reinforcement learning problem, where the state space is represented by the auction information and the campaign's real-time parameters, while an action is the bid price to set.
11 code implementations • 1 Nov 2016 • Yanru Qu, Han Cai, Kan Ren, Wei-Nan Zhang, Yong Yu, Ying Wen, Jun Wang
Predicting user responses, such as clicks and conversions, is of great importance and has found its usage in many Web applications including recommender systems, web search and online advertising.
Ranked #1 on Click-Through Rate Prediction on iPinYou