@Innov_Inf profile picture
The Innovation | Informatics @Innov_Inf

Artificial intelligence research within reach. The official account of The Innovation | Informatics. Tweets by @YangLiu_CTH. #AI #NEWS #Trending #ML

Joined October 2021
  • Tweets 3,348
  • Following 4,488
  • Followers 3,684
1,894 Photos and videos
Spatial-Temporal Identity: A Simple yet Effective Baseline for Multivariate Time Series Forecasting A promising method nearly 200 stars, and used as baselines by nearly 100 papers. github.com/GestaltCogTeam/ST… #timeseries #MachineLearning #deeplearning #artificalintelligence #bigdata
1
157
Spatial-Temporal-Decoupled Masked Pre-training for Spatiotemporal Forecasting arxiv.org/abs/2312.00516 #ChatGPT #Timeseries #datascience #artificalintelligence #Prediction #deeplearning #MachineLearning #STGNN #bigdata
2
1
121
UniST: A Prompt-Empowered Universal Model for Urban Spatio-Temporal Prediction dl.acm.org/doi/abs/10.1145/3… #ChatGPT #Timeseries #datascience #urbancomputing #artificalintelligence #Prediction #deeplearning #MachineLearning #STGNN #bigdata @Tsinghua_Uni
1
1
109
Common occurrences of subsurface heatwaves and cold spells in ocean eddies nature.com/articles/s41586-0… #Science #heatwaves #timeseries #ocean
87
Chronos: Learning the Language of Time Series arxiv.org/abs/2403.07815 #ChatGPT #Timeseries #datascience #artificalintelligence #Prediction #deeplearning #MachineLearning #STGNN #bigdata @amazon

Chronos: Learning the Language of Time Series

We introduce Chronos, a simple yet effective framework for pretrained probabilistic time series models. Chronos tokenizes time series values using scaling and quantization into a fixed vocabulary...

arxiv.org
2
93
The Innovation | Informatics retweeted
Xiaoxiao Wang@xiaoxiao_mri
15 Nov 2024
This's why we insist in whole-brain decoding. "behavioral signatures can be decoded from a much broader range of cortical areas than previously recognized" direct.mit.edu/imag/article/…
1
19
117
17,177
Pre-training enhanced spatial-temporal graph neural network for multivariate time series forecasting dl.acm.org/doi/abs/10.1145/3… #ChatGPT #Timeseries #datascience #artificalintelligence #Prediction #deeplearning #MachineLearning #STGNN #bigdata
2
3
156
BigST: Linear Complexity Spatio-Temporal Graph Neural Network for Traffic Forecasting on Large-Scale Road Networks dl.acm.org/doi/10.14778/3641… #ChatGPT #Timeseries #datascience #artificalintelligence #Prediction #deeplearning #MachineLearning #STGNN #bigdata

BigST: Linear Complexity Spatio-Temporal Graph Neural Network for Traffic Forecasting on Large-Sc...

Spatio-Temporal Graph Neural Network (STGNN) has been used as a common workhorse for traffic forecasting. However, most of them require prohibitive quadratic computational complexity to capture...

dl.acm.org
1
3
124
The Innovation | Informatics retweeted
Shehzaad Dhuliawala@shehzaadzd
15 Nov 2024
Excited to share work from my internship at @AIatMeta! LLM devs often tweak decoding temperature: low for analytical tasks, and high for creative ones. Why not learn this from the data? Introducing the AdaptiveDecoder! (1/3)🧵
Jason Weston
@jaseweston
15 Nov 2024
🚨 Adaptive Decoding via Latent Preference Optimization 🚨 - New layer added to Transformer, selects decoding params automatically *per token* - Learnt via new method, Latent Preference Optimization - Outperforms any fixed temperature decoding method, choosing creativity or factuality automatically arxiv.org/abs/2411.09661 🧵1/4
4
29
256
31,641
The Innovation | Informatics retweeted
anandmaj
@Almondgodd
14 Nov 2024
I trained a 2-arm robot using my Nintendo Switch ( an action transformer) how I did it my most interesting takeaways below👇🏼
0:26
41
142
1,214
136,269
The Innovation | Informatics retweeted
merve
@mervenoyann
14 Nov 2024
Microsoft released LLM2CLIP: a CLIP model with longer context window for complex text inputs 🤯 TLDR; they replaced CLIP's text encoder with various LLMs fine-tuned on captioning, better top-k accuracy on retrieval 🔥 All models with Apache 2.0 license on @huggingface 😍
6
111
657
36,804
The Innovation | Informatics retweeted
Valeriy M., PhD, MBA, CQF
@predict_addict
15 Nov 2024
A new promising “A Fair and Scalable Time Series Forecasting Benchmark and Toolkit” BasicTS. #timeseries #forecasting #opensource
3
24
157
6,735
The Innovation | Informatics retweeted
Valeriy M., PhD, MBA, CQF
@predict_addict
15 Nov 2024
Forget traditional methods like SMOTE—Conformal Prediction offers significant improvements, even on challenging datasets with imbalanced data, label noise, or both. 🚀 #conformalprediction #MachineLearning #DataScience #Innovation
2
10
72
3,229
The Innovation | Informatics retweeted
elvis
@omarsar0
14 Nov 2024
Microsoft released a neat Python library to run LLM-powered multi-agent simulations of people with personalities, interests, and goals.
19
205
1,221
84,457
The Innovation | Informatics retweeted
Science News@SciencNews
14 Nov 2024
How to write an excellent Review article @natrevbioeng @Nature #PhD #PhDlife #academics #AcademicChatter #AcademicTwitter #Postdocs #Research #Science
7
644
2,548
431,870
The Innovation | Informatics retweeted
Christian S. Perone@tarantulae
14 Nov 2024
New article: "The geometry of data: the missing metric tensor and the Stein score" (blog.christianperone.com/202…). I show how you can derive a (efficient to compute) data manifold metric tensor with the Stein score alone ! Deep connections to diffusion, score-based models and physics.
14
116
836
60,919
The Innovation | Informatics retweeted
The Innovation: a Cell Press partner journal
@The_InnovationJ
15 Nov 2024
New in The Innovation Geoscience! Larger increase in future global terrestrial water availability than projected by CMIP6 models. Wu et al. employ outputs from 20 state-of-the-art ESMs from the Coupled Model Intercomparison Project Phase 6, together with multiple remote sensing and ground-based precipitation and ET datasets to derive new projections of ΔPME under shared socioeconomic pathway 2-4.5 and 5-8.5 scenarios using the EC approach. Read more @Innov_Geosci the-innovation.org/article/d… #geoscience #Research #climate
1
1
411
The Innovation | Informatics retweeted
Science Magazine
@ScienceMagazine
14 Nov 2024
A new Science study presents “Evo”—a machine learning model capable of decoding and designing DNA, RNA, and protein sequences, from molecular to genome scale, with unparalleled accuracy. Evo’s ability to predict, generate, and engineer entire genomic sequences could change the way synthetic biology is done. Learn more in this week's issue: bit.ly/3OsmUPr
53
412
1,258
882,208
The Innovation | Informatics retweeted
iMeta@iMetaScience
15 Nov 2024
#Multiomics integration unravels #genotype‐#microbiome interactions shaping the conjunctival transcriptome #iMetaOmics onlinelibrary.wiley.com/doi/… #bioinformatics #DataScience #MachineLearning @wileymicrobio @WileyBiomedical @WileyHealth
23
16
2,417
Artificial intelligence for geoscience: Progress, challenges, and perspectives cell.com/the-innovation/full… #ArtificialIntelligence #MachineLearning #Geoscience #Science #Research #Trending #PERSPECTIVE #AI #Technology
1
2
129
Load more