no code implementations • 16 Apr 2024 • Xiao Zhang, Ruoxi Jiang, William Gao, Rebecca Willett, Michael Maire
We demonstrate that adding a weighting factor to decay the strength of identity shortcuts within residual networks substantially improves semantic feature learning in the state-of-the-art self-supervised masked autoencoding (MAE) paradigm.
no code implementations • 11 Dec 2023 • Xiao Zhang, David Yunis, Michael Maire
We present an approach for analyzing grouping information contained within a neural network's activations, permitting extraction of spatial layout and semantic segmentation from the behavior of large pre-trained vision models.
no code implementations • 23 Nov 2023 • Cyrus Zhou, Pedro Savarese, Vaughn Richard, Zack Hassman, Xin Yuan, Michael Maire, Michael DiBrino, Yanjing Li
We present an end-to-end co-design approach encompassing computer architecture, training algorithm, and inference optimization to efficiently execute networks with fine-grained heterogeneous precisions.
no code implementations • 26 Oct 2023 • Sudarshan Babu, Richard Liu, Avery Zhou, Michael Maire, Greg Shakhnarovich, Rana Hanocka
We introduce HyperFields, a method for generating text-conditioned Neural Radiance Fields (NeRFs) with a single forward pass and (optionally) some fine-tuning.
1 code implementation • 11 Oct 2023 • YuHan Liu, Hanchen Li, Yihua Cheng, Siddhant Ray, YuYang Huang, Qizheng Zhang, Kuntai Du, Jiayi Yao, Shan Lu, Ganesh Ananthanarayanan, Michael Maire, Henry Hoffmann, Ari Holtzman, Junchen Jiang
Compared to the recent systems that reuse the KV cache, CacheGen reduces the KV cache size by 3. 5-4. 3x and the total delay in fetching and processing contexts by 3. 2-3. 7x while having negligible impact on the LLM response quality in accuracy or perplexity.
no code implementations • 7 Oct 2023 • YuHan Liu, Chengcheng Wan, Kuntai Du, Henry Hoffmann, Junchen Jiang, Shan Lu, Michael Maire
ML APIs have greatly relieved application developers of the burden to design and train their own neural network models -- classifying objects in an image can now be as simple as one line of Python code to call an API.
1 code implementation • 30 Sep 2023 • Xiao Zhang, Michael Maire
Within the framework of generative adversarial networks (GANs), we propose objectives that task the discriminator for self-supervised representation learning via additional structural modeling responsibilities.
no code implementations • 27 Sep 2023 • Xin Yuan, Michael Maire
We develop a neural network architecture which, trained in an unsupervised manner as a denoising diffusion model, simultaneously learns to both generate and segment images.
no code implementations • 31 May 2023 • Zhuokai Zhao, Takumi Matsuzawa, William Irvine, Michael Maire, Gordon L Kindlmann
NERO evaluation is consist of a task-agnostic interactive interface and a set of visualizations, called NERO plots, which reveals the equivariance property of the model.
no code implementations • NeurIPS 2021 • Sudarshan Babu, Pedro Savarese, Michael Maire
We demonstrate that efficient meta-learning can be achieved via end-to-end training of deep neural networks with memory distributed across layers.
no code implementations • CVPR 2021 • Xin Yuan, Zhe Lin, Jason Kuen, Jianming Zhang, Yilin Wang, Michael Maire, Ajinkya Kale, Baldo Faieta
We first train our model on COCO and evaluate the learned visual representations on various downstream tasks including image classification, object detection, and instance segmentation.
1 code implementation • CVPR 2021 • Pedro Savarese, Sunnie S. Y. Kim, Michael Maire, Greg Shakhnarovich, David Mcallester
We study image segmentation from an information-theoretic perspective, proposing a novel adversarial method that performs unsupervised segmentation by partitioning images into maximally independent sets.
Ranked #1 on Unsupervised Image Segmentation on Flowers
no code implementations • NeurIPS 2020 • Xiao Zhang, Michael Maire
We create a framework for bootstrapping visual representation learning from a primitive visual grouping capability.
1 code implementation • 23 Nov 2020 • Tri Huynh, Simon Kornblith, Matthew R. Walter, Michael Maire, Maryam Khademi
While positive pairs can be generated reliably (e. g., as different views of the same image), it is difficult to accurately establish negative pairs, defined as samples from different images regardless of their semantic content or visual features.
no code implementations • ICML 2020 • Chengcheng Wan, Henry Hoffmann, Shan Lu, Michael Maire
We propose a novel variant of SGD customized for training network architectures that support anytime behavior: such networks produce a series of increasingly accurate outputs over time.
no code implementations • ICLR 2021 • Xin Yuan, Pedro Savarese, Michael Maire
We develop an approach to growing deep network architectures over the course of training, driven by a principled combination of accuracy and sparsity objectives.
no code implementations • CVPR 2020 • Haochen Wang, Ruotian Luo, Michael Maire, Greg Shakhnarovich
The core of our approach, Pixel Consensus Voting, is a framework for instance segmentation based on the Generalized Hough transform.
Ranked #36 on Panoptic Segmentation on COCO test-dev
2 code implementations • NeurIPS 2020 • Pedro Savarese, Hugo Silva, Michael Maire
Additionally, the recent Lottery Ticket Hypothesis conjectures that, for a typically-sized neural network, it is possible to find small sub-networks which, when trained from scratch on a comparable budget, match the performance of the original dense counterpart.
1 code implementation • CVPR 2021 • Pedro Savarese, David Mcallester, Sudarshan Babu, Michael Maire
From a simplified analysis of adaptive methods, we derive AvaGrad, a new optimizer which outperforms SGD on vision tasks when its adaptability is properly tuned.
no code implementations • 31 Oct 2019 • Chengcheng Wan, Muhammad Santriaji, Eri Rogers, Henry Hoffmann, Michael Maire, Shan Lu
An increasing number of software applications incorporate runtime Deep Neural Networks (DNNs) to process sensor data and return inference results to humans.
1 code implementation • ICML 2020 • Tri Huynh, Michael Maire, Matthew R. Walter
We introduce a novel approach to endowing neural networks with emergent, long-term, large-scale memory.
1 code implementation • ICLR 2019 • Pedro Savarese, Michael Maire
Restricting the number of templates yields a flexible hybridization of traditional CNNs and recurrent networks.
no code implementations • CVPR 2018 • Mohammadreza Mostajabi, Michael Maire, Gregory Shakhnarovich
Our technique is applicable when the ground-truth labels themselves exhibit internal structure; we derive a regularizer by learning an autoencoder over the set of annotations.
2 code implementations • ECCV 2018 • Ligeng Zhu, Ruizhi Deng, Michael Maire, Zhiwei Deng, Greg Mori, Ping Tan
We explore a key architectural aspect of deep convolutional neural networks: the pattern of internal skip connections used to aggregate outputs of earlier layers for consumption by deeper layers.
no code implementations • ECCV 2018 • Huaizu Jiang, Erik Learned-Miller, Gustav Larsson, Michael Maire, Greg Shakhnarovich
As an agent moves through the world, the apparent motion of scene elements is (usually) inversely proportional to their depth.
1 code implementation • CVPR 2017 • Gustav Larsson, Michael Maire, Gregory Shakhnarovich
How many labels are needed?
1 code implementation • CVPR 2017 • Tsung-Wei Ke, Michael Maire, Stella X. Yu
Most critically, multigrid structure enables networks to learn internal attention and dynamic routing mechanisms, and use them to accomplish tasks on which modern CNNs fail.
4 code implementations • 24 May 2016 • Gustav Larsson, Michael Maire, Gregory Shakhnarovich
We introduce a design strategy for neural network macro-architecture based on self-similarity.
Ranked #29 on Image Classification on SVHN
3 code implementations • 22 Mar 2016 • Gustav Larsson, Michael Maire, Gregory Shakhnarovich
This intermediate output can be used to automatically generate a color image, or further manipulated prior to image formation.
no code implementations • CVPR 2016 • Michael Maire, Takuya Narihira, Stella X. Yu
Spectral embedding provides a framework for solving perceptual organization problems, including image segmentation and figure/ground organization.
no code implementations • ICCV 2015 • Takuya Narihira, Michael Maire, Stella X. Yu
We demonstrate results on both the synthetic images of Sintel and the real images of the classic MIT intrinsic image dataset.
no code implementations • CVPR 2015 • Takuya Narihira, Michael Maire, Stella X. Yu
We develop a new approach to inferring lightness, the perceived reflectance of surfaces, from a single image.
no code implementations • 16 Oct 2014 • Michael Maire, Stella X. Yu, Pietro Perona
We frame the task of predicting a semantic labeling as a sparse reconstruction procedure that applies a target-specific learned transfer function to a generic deep sparse code representation of an image.
35 code implementations • 1 May 2014 • Tsung-Yi Lin, Michael Maire, Serge Belongie, Lubomir Bourdev, Ross Girshick, James Hays, Pietro Perona, Deva Ramanan, C. Lawrence Zitnick, Piotr Dollár
We present a new dataset with the goal of advancing the state-of-the-art in object recognition by placing the question of object recognition in the context of the broader question of scene understanding.