#TransferLearning
LightNews LightNews
ossradarai.bsky.social
A new arXiv method adapts frozen flow-based VLA policies like π₀ to new robot embodiments without demonstrations by using a lightweight classifier to guide denoising steps, showing gains on the LIBERO benchmark.

#Robotics #OpenSourceAI #VLAPolicies #TransferLearning
https://arxiv.org/abs/2610.22576
October 19, 2026 at 6:01 AM
ai-bloom.warp-studio.com
MIT開発のAI「IRIS」：細胞間シグナリング経路の普遍的フィンガープリント解読と転移学習

MITが開発したニューラルネットワークモデル「IRIS」により、細胞間コミュニケーションのシグナリング履歴を高精度に再構築が可能に。

#BioAI #MachineLearning #NeuralNetworks #TransferLearning #CellBiology
MIT開発のAI「IRIS」：細胞間シグナリング経路の普遍的フィンガープリント解読と転移学習
MITが開発したニューラルネットワークモデル「IRIS」により、細胞間コミュニケーションのシグナリング履歴を高精度に再構築が可能に。
ai.warp-studio.com
October 16, 2026 at 4:11 PM
sergiocuellar.bsky.social
Transfer learning with PyTorch: A frozen ResNet18 trained on 30 images reached 97.78% accuracy, while training from scratch only hit 33.33%. Normalizing input images is crucial for the model to perform accurately. #TransferLearning #PyTorch #MachineLearning
Transfer Learning in PyTorch Using a Pretrained ResNet Model
Training 1,539 parameters beat training 11.2 million, on the same data, in less time. This walkthrough shows exactly how transfer learning achieves that in PyTorch, with both experiments run side by side and every number measured rather than claimed.
kodekloud.com
September 23, 2026 at 4:39 AM
gp-pulipaka.bsky.social
#TransferLearning in #MachineLearning with #Tensorflow Examples. #BigData #Analytics #DataScience #AI #IoT #IIoT #Python #RStats #Java #JavaScript #ReactJS #CloudComputing #Serverless #DataScientist #Linux #Statistics #Programming #Coding #100DaysofCode
geni.us/TL-Example
September 21, 2026 at 6:42 AM
alecrimi.bsky.social
Here comes the dream-reader from just #EEG:
"Interpreting human sleep activity through neural contrastive learning"
www.cell.com/neuron/fullt...
#Deeplearning #transferlearning #brain
September 3, 2026 at 11:22 AM
jpecol.bsky.social
【🎉Latest accepted article】
Field estimation of leaf water potential in poplar trees using UAV-based hyperspectral imagery and #DeepLearning

#Cross_scaleModeling | #ForestDroughtMonitoring | #HyperspectralRemoteSensing | #TransferLearning

@mapjournals.bsky.social

doi.org/10.1093/jpe/...
July 2, 2026 at 2:14 PM
zidan-jaffar.bsky.social
Pretrained models are like mentors.
You learn faster when someone smarter shows you the way first.

#DeepLearning #TransferLearning
July 25, 2025 at 3:33 PM
mbeyeler.bsky.social
We introduce BIRD: Behavior Induction via Representation-structure Distillation.

Instead of transferring outputs, BIRD aligns the geometry of internal representations between teacher and student, enabling weak → strong generalization.

#KnowledgeDistillation #TransferLearning #Robustness
Two-panel schematic illustrating the BIRD framework. Left panel shows independent pre-training of a teacher and a student network on different datasets, each optimized with its own task loss. Right panel shows representation-structure distillation: selected intermediate layers from teacher and student are compared via a representation loss, which aligns the geometry of their internal activations while the student is still trained on its own task loss. A snowflake icon indicates the teacher is frozen. The diagram emphasizes that behavior is transferred by aligning internal representation structure rather than outputs or shared data.
February 8, 2026 at 5:35 PM
bbloggsbott.bsky.social
Just published my deep dive into the T5 paper

T5's big insight? Treat EVERY NLP task as text-to-text:
- Translation
- Classification
- Even regression

vishalramesh.com/read-a-paper...
#MachineLearning #NLP #AI #TransferLearning
Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer (T5)
Dive deep into the ‘Attention Is All You Need’ paper. This detailed guide deconstructs the Transformer architecture, explaining Self-Attention, Multi-Head Attention, and how it replaced RNNs in NLP.
vishalramesh.com
December 27, 2025 at 4:13 PM
ravr19.bsky.social
3/3 Transfer learning (TL) in soil spectroscopy: Our framework adapts domain similarities (homogeneous/heterogeneous) and TL categories (inductive, transductive, unsupervised). TL is underexplored in soil spectroscopy—a promising area for research! 🔬📊 #SoilScience #TransferLearning #SLSL
classification of transfer learning (TL). Dashed lines extending from Heterogeneous TL indicate that we have found no studies on heterogeneous TL in the soil spectroscopic literature.
November 23, 2024 at 8:45 AM
ravr19.bsky.social
2/3 This figure highlights methods for localizing soil spectroscopic modeling:
🌍 Global models: Broad but often biased.
📍 Local models: Accurate but costly.
Localization bridges the gap, enhancing accuracy & cost-effectiveness in soil spectroscopy. 🌈📊 #SoilScience #TransferLearning #SLSL
methods for localizing soil spectroscopic modelling. Models built using all observations in a soil spectral library (SSL; Global modelling) tend to be biased. Models built using only local observations (Local modelling) are accurate but expensive. Localization methods aim to improve the accuracy and cost-effectiveness of soil spectroscopy
November 23, 2024 at 8:44 AM
ravr19.bsky.social
1/3 🌍 Soil spectroscopy meets transfer learning! Global spectral libraries (SSLs) offer insights but lack local precision. Transferring knowledge enhances predictions. Must think globally but fit locally! 🌈📊🌱 #SoilScience #MachineLearning #TransferLearning #SLSL. 🔗 Read more: tinyurl.com/mrva8vt2
November 23, 2024 at 8:43 AM
scienceopen.bsky.social
'Comprehensive Evaluation of Pre-trained Neural Networks for Alzheimer’s Disease Classification Using Transfer Learning' - a Journal of #DisabilityResearch article in the #KSCDR Collection on #ScienceOpen:

🖇️ #AlzheimersDisease #MRI #TransferLearning #MedicalImaging
Comprehensive Evaluation of Pre-trained Neural Networks for Alzheimer’s Disease Classification Using Transfer Learning
Alzheimer’s disease (AD) is a progressive neurodegenerative disorder that requires early and accurate diagnosis for effective intervention.
www.scienceopen.com
September 9, 2025 at 2:54 PM
iopp-mlresearch.bsky.social
In this #MachineLearningScienceandTechnology article (with a sprinkle of @iopp-quantum.bsky.social), Sgroi and coworkers design a chain of particles that achieves efficient excitation-transfer performances using a #transferlearning approach. Read it here: iopscience.iop.org/article/10.1...
Figure 2 of the article, showing a 'Sketch of the physical problem: two mutually interacting particles, A and B, undergo excitation-transfer processes from A, initially prepared in its excited state, to B. The latter is also incoherently coupled to a sink. We consider this as a starting configuration for the building up of a network: one by one, we add particles between A and B and seek for the potential increase of the excitation-transfer efficiency. The inter-particle interactions depends on their relative positions.'
May 8, 2025 at 4:59 PM
splashking.bsky.social
I am very happy we finally publish our paper on transfer learning in Smart Cities in the prestigious "International Journal of Geographical Information Science". #transferlearning #missingdata #geospatialai #airquality #smartcity

www.tandfonline.com/doi/full/10....
Smart Beestricts: improving the spatial resolution of air-quality data in Madrid through transfer learning
Sensor infrastructures have become key enablers in collecting massive urban data. While representing an invaluable source of geographic information, high deployment costs of air quality stations re...
www.tandfonline.com
May 7, 2025 at 6:49 AM
transferlearning.bsky.social
Transferlearning to fight wildfires such Palisades fire ?
January 15, 2025 at 2:36 PM
transferlearning.bsky.social
catch Transferlearning waves!
December 12, 2024 at 7:15 AM
envdatascience.bsky.social
Recent article!

Time series predictions in unmonitored sites: a survey of #MachineLearning techniques in #water #resources

doi.org/10.1017/eds....

Jared D. Willard and colleagues @berkeleylab.lbl.gov, University of Pittsburgh & University of Minnesota Twin Cities

#DeepLearning #TransferLearning
February 5, 2025 at 2:35 PM
stephaniehicks.bsky.social
Congratulations to @findycang.bsky.social from @jhubiostat.bsky.social who did an amazing job presenting her work at #biodata24 tonight! Preprint coming soon. ☺️

#ProudPI #stats #spatial #omics #transferlearning 🧪🧬🖥️🧠📈
Cindy Fang presents her work at CSHL Biological Data Science
November 15, 2024 at 4:07 AM
arbisoft.bsky.social
Imagine if you could transfer intelligence...
With AI, you can! 🧠
Transfer learning helps you skip the grunt work and go straight to genius.

Abdul Moiz, explains how pre-trained models are saving time: https://f.mtr.cool/wklubmvytp

#ArbisoftBlogs #TransferLearning #AI
Transfer Learning in Deep Learning: Using Pre-trained Models for Quick Wins
Speed up AI projects with transfer learning! Learn how pre-trained models slash training time, boost accuracy, and deliver quick wins. Master practical strategies for deep learning success.
f.mtr.cool
May 28, 2025 at 5:31 AM
arbisoft.bsky.social
Build faster. Learn smarter.
Transfer learning lets you skip to the good part with models that already know the ropes.
👉 Read the blog by Abdul Moiz, Software Developer at Arbisoft: https://f.mtr.cool/klutsijvif

#TransferLearning #DeepLearning #AI #Arbisoft #MachineLearning
May 6, 2025 at 2:03 PM
ml-nn-eu.bsky.social
Transfer learning is a technique in machine learning where a model developed for one task is reused as the starting point for a model on a second task. Rather than training a model entirely from scratch[..]

Full article:
ml-nn.eu/a1/86.html

#machinelearning #ai #transferlearning
Transfer Learning in Machine Learning
Machine Learning & Neural Networks Blog
ml-nn.eu
February 16, 2026 at 4:22 PM
puretech.news
Building voice AI that listens to everyone: Transfer learning and synthetic speech in action #Technology #EmergingTechnologies #ArtificialIntelligence #VoiceAI #TransferLearning #SyntheticSpeech
Building voice AI that listens to everyone: Transfer learning and synthetic speech in action
Enterprises adopting voice AI must consider not just usability, but inclusion. Supporting users with disabilities is a market opportunity.Read More
puretech.news
July 12, 2025 at 10:00 PM
getnews-me.bsky.social
Research accepted to NeurIPS 2025 finds that larger per‑class fine‑tuning datasets lower membership inference attack advantage via a power‑law relationship. Read more: https://getnews.me/dataset-size-shapes-membership-inference-risk-in-transfer-learning/ #membershipinference #transferlearning
Dataset Size Shapes Membership Inference Risk in Transfer Learning
October 8, 2025 at 11:57 AM
getnews-me.bsky.social
Transfer learning reduced mean absolute percentage error by 10.6% for sister vessels and 5.3% for different vessels, outperforming models built only on noon reports. Read more: https://getnews.me/transfer-learning-improves-maritime-shaft-power-forecasts/ #transferlearning #maritime #shaftpower
Transfer Learning Improves Maritime Shaft Power Forecasts
October 6, 2025 at 9:35 AM