NEW BOT Телеграм, страница

🖥

CANIFE: Crafting Canaries for Empirical Privacy Measurement in Federated Learning

A novel method, CANIFE, that uses canaries - carefully crafted samples by a strong adversary to evaluate the empirical privacy of a training round.

conda create -n "canife" python=3.9 
conda activate canife
pip install -r ./requirements.txt

⚙️Github: https://github.com/facebookresearch/canife

📄Paper: https://arxiv.org/abs/2210.02912v1

🗒Dataset: https://paperswithcode.com/dataset/cifar-10

@ArtificialIntelligencedl

Please open Telegram to view this post

VIEW IN TELEGRAM

👍6❤1🔥1

1.79K views08:15

Artificial Intelligence

💻

A Closer Look at Hardware-Friendly Weight Quantization

⚙️

Github: https://github.com/google/qkeras

📄Paper: https://arxiv.org/abs/2210.03671v1

🗒Training: https://github.com/BertMoons/QuantizedNeuralNetworks-Keras-Tensorflow

@ArtificialIntelligencedl

Please open Telegram to view this post

VIEW IN TELEGRAM

👍5❤1🔥1

1.75K views07:38

Artificial Intelligence

🗄

NerfAcc: A General NeRF Acceleration Toolbox

pip install nerfacc

⚙️Github: https://github.com/kair-bair/nerfacc

📄Paper: https://arxiv.org/abs/2210.04847v1

🗒Dataset: https://paperswithcode.com/dataset/nerf

@ArtificialIntelligencedl

Please open Telegram to view this post

VIEW IN TELEGRAM

GitHub

GitHub - nerfstudio-project/nerfacc: A General NeRF Acceleration Toolbox in PyTorch.

A General NeRF Acceleration Toolbox in PyTorch. Contribute to nerfstudio-project/nerfacc development by creating an account on GitHub.

👍5❤1🔥1

1.87K viewsedited 08:52

Artificial Intelligence

👣

OGC: Unsupervised 3D Object Segmentation from Rigid Dynamics of Point Clouds

⚙️Github: https://github.com/vlar-group/ogc

📄Paper: https://arxiv.org/abs/2210.04458v1

↪️

Demo: https://www.youtube.com/watch?v=dZBjvKWJ4K0

🗒Dataset: https://paperswithcode.com/dataset/kitti

@ArtificialIntelligencedl

Please open Telegram to view this post

VIEW IN TELEGRAM

👍3❤1🔥1

1.65K views08:40

Artificial Intelligence

🛠

Understanding Embodied Reference with Touch-Line Transformer

conda create --name nvvc python=3.8
conda activate nvvc
pip install -r requirements.txt
conda install pytorch torchvision torchaudio cudatoolkit=11.3 -c pytorch

⚙️Github: https://github.com/yang-li-2000/understanding-embodied-reference-with-touch-line-transformer

📄Paper: https://arxiv.org/abs/2210.05668v2

🗒Dataset: https://paperswithcode.com/dataset/refcoco

@ArtificialIntelligencedl

Please open Telegram to view this post

VIEW IN TELEGRAM

👍4❤1🔥1

1.65K viewsedited 08:25

Artificial Intelligence

🖥

Exploring Long-Sequence Masked Autoencoders

⚙️Github: https://github.com/facebookresearch/long_seq_mae

📄Paper: https://arxiv.org/abs/2210.07224v1

🗒Dataset: https://paperswithcode.com/dataset/places

@ArtificialIntelligencedl

Please open Telegram to view this post

VIEW IN TELEGRAM

👍4❤1🔥1

1.54K views08:07

Artificial Intelligence

🗄

SUPERB-prosody: On The Utility of Self-supervised Models for Prosody-related Tasks

⚙️Github: https://github.com/jsalt-2022-ssl/superb-prosody

📄Paper: https://arxiv.org/abs/2210.07185v1

🗒Tasks: https://paperswithcode.com/task/prosody-prediction

@ArtificialIntelligencedl

Please open Telegram to view this post

VIEW IN TELEGRAM

👍6❤1🔥1

1.53K views06:39

Artificial Intelligence

🖥

Model-Based Offline Reinforcement Learning with Pessimism-Modulated Dynamics Belief

⚙️Github: https://github.com/huawei-noah/HEBO/tree/master/PMDB

📄Paper: https://arxiv.org/abs/2210.06692v1

↪️

Project: https://research.nianticlabs.com/mapfree-reloc-benchmark

🗒Dataset: https://paperswithcode.com/dataset/d4rl

@ai_machinelearning_big_data

Please open Telegram to view this post

VIEW IN TELEGRAM

GitHub

HEBO/PMDB at master · huawei-noah/HEBO

Bayesian optimisation & Reinforcement Learning library developped by Huawei Noah's Ark Lab - huawei-noah/HEBO

👍7❤2🔥1

1.68K views07:02

Artificial Intelligence

🔩

SubeventWriter: Iterative Sub-event Sequence Generation with Coherence Controller

⚙️Github: https://github.com/hkust-knowcomp/subeventwriter

📄Paper: https://arxiv.org/abs/2210.06694v1

🗒Dataset: https://paperswithcode.com/dataset/wikihow

@ArtificialIntelligencedl

Please open Telegram to view this post

VIEW IN TELEGRAM

👍4❤1🔥1

1.63K viewsedited 07:58

Artificial Intelligence

🖥

Legal Case Document Summarization: Extractive and Abstractive Methods and their Evaluation

⚙️Github: https://github.com/law-ai/summarization

📄Paper: https://arxiv.org/pdf/2210.07544v1.pdf

🗒Dataset: http://www.liiofindia.org/in/cases/cen/INSC/

@ArtificialIntelligencedl

Please open Telegram to view this post

VIEW IN TELEGRAM

GitHub

GitHub - Law-AI/summarization: Implementation of different summarization algorithms applied to legal case judgements.

Implementation of different summarization algorithms applied to legal case judgements. - GitHub - Law-AI/summarization: Implementation of different summarization algorithms applied to legal case ju...

👍6❤1🔥1

1.69K views07:42

Artificial Intelligence

✅

GNN-Motion-Planning

⚙️Github: https://github.com/rainorangelemon/gnn-motion-planning

📄Paper: https://arxiv.org/abs/2210.08864v1

🗒Project: https://rainorangelemon.github.io/NeurIPS2021/

@ArtificialIntelligencedl

Please open Telegram to view this post

VIEW IN TELEGRAM

👍8❤1🥰1

1.55K views12:15

Artificial Intelligence

🖥

RibSeg v2: A Large-scale Benchmark for Rib Labeling and Anatomical Centerline Extraction

⚙️Github: https://github.com/m3dv/ribseg

📄Paper: https://arxiv.org/abs/2210.09309v1

🗒Dataset: https://doi.org/10.5281/zenodo.5336592

@ArtificialIntelligencedl

Please open Telegram to view this post

VIEW IN TELEGRAM

❤3👍1🔥1

1.53K viewsedited 06:26

Artificial Intelligence

🖥

HyperDomainNet: Universal Domain Adaptation for Generative Adversarial Networks (NeurIPS 2022)

🖥

Github: https://github.com/macderru/hyperdomainnet

📄Paper: https://arxiv.org/abs/2210.08884v2

🔩

Colab: https://colab.research.google.com/drive/1QMylWjzPxvHtxm74U4lWRQXwquw5AaFL#scrollTo=si2tLKYLT-kV

@ArtificialIntelligencedl

Please open Telegram to view this post

VIEW IN TELEGRAM

👍5❤1🔥1

1.67K views10:19

Artificial Intelligence

↪️

Targeted Adversarial Self-Supervised Learning

🖥

Github: https://github.com/Kim-Minseon/RoCL

📄Paper: https://arxiv.org/abs/2210.10482v1

🔩

Adversarial Self-Supervised Contrastive Learning: https://sites.google.com/view/rocl2020

@ArtificialIntelligencedl

Please open Telegram to view this post

VIEW IN TELEGRAM

❤4👍2🔥2

1.65K views07:51

Artificial Intelligence

⏩

Learning to Discover and Detect Objects

🖥

Github: https://github.com/vlfom/rncdl

➡️

Project: https://vlfom.github.io/RNCDL/

📄Paper: https://arxiv.org/abs/2210.10774v1

🔩

Dataset: https://paperswithcode.com/dataset/visual-genome

@ArtificialIntelligencedl

Please open Telegram to view this post

VIEW IN TELEGRAM

❤6🔥4👍2

1.59K views06:38

Artificial Intelligence

⭐️

Multi-hypothesis 3D human pose estimation metrics favor miscalibrated distributions

🖥

Github: https://github.com/sinzlab/cgnf

➡️

Model: https://github.com/sinzlab/propose/tree/0.2.0/propose/models/flows

📄Paper: https://arxiv.org/abs/2210.11179v1

🔩

Dataset: https://paperswithcode.com/dataset/human3-6m

@ArtificialIntelligencedl

Please open Telegram to view this post

VIEW IN TELEGRAM

👍5❤1🔥1

1.64K views07:18

Artificial Intelligence

🖥

TOIST: Task Oriented Instance Segmentation Transformer with Noun-Pronoun Distillation

🖥

Github: https://github.com/air-discover/toist

📄Paper: https://arxiv.org/abs/2210.10775v1

🔩

Dataset: https://github.com/coco-tasks/dataset

@ArtificialIntelligencedl

Please open Telegram to view this post

VIEW IN TELEGRAM

👍5❤1🔥1🤯1

1.59K views07:52

Artificial Intelligence

Reversed Image Signal Processing and RAW Reconstruction. AIM 2022 Challenge Report

🖥

Github: https://github.com/mv-lab/AISP

📄Paper: https://arxiv.org/abs/2210.11153v1

🔩

Starter guide: https://github.com/mv-lab/AISP/blob/main/aim22-reverseisp/official-starter-code.ipynb

@ArtificialIntelligencedl

Please open Telegram to view this post

VIEW IN TELEGRAM

👍4🔥2❤1

1.56K views06:29

Artificial Intelligence

🖥

NVIDIA Federated Learning Application Runtime Environment

NVIDIA FLARE enables researchers to collaborate and build AI models without sharing private data.

pip install nvflare

🖥

Github: https://github.com/nvidia/nvflare

📄Paper: https://arxiv.org/abs/2210.13291v1

🔩

Starter guide: https://nvflare.readthedocs.io/en/main/getting_started.html

@ArtificialIntelligencedl

Please open Telegram to view this post

VIEW IN TELEGRAM

👍9🤔2

1.54K views06:05

Artificial Intelligence

🖥

Evaluating Long-Term Memory in 3D Mazes

pip install memory-maze

🖥