雁过留声
4天前
雁过留声
4天前
The Figen
1周前
Yann LeCun
1周前
OpenAI
1周前
Wild Clips
1周前
Wild Clips
1周前
勃勃OC
2周前
NO CONTEXT HUMANS
2周前
NO CONTEXT HUMANS
2周前
The Spectator Index
2周前
🇺🇸 USS GERALD R. FORD - Lead aircraft carrier of its class and the biggest warship ever built. - Cost of $13 billion, making it the most expensive warship in history. - Length of 1,092 ft (333 m) and displacement of 100,000 tonnes. - Powered by two nuclear reactors, with operational range of 25 years before a mid-life refuel is needed. - Speed in excess of 30 knots (56 km/h or 35 mph). - Armed with its own surface-to-air missiles and guns. - Can carry over 75 military aircraft, including F-35 fighter jets.
Wild Clips
2周前
NO CONTEXT HUMANS
2周前
Isaac Stone Fish
2周前
NO CONTEXT HUMANS
2周前
The Spectator Index
2周前
Leo Xiang
2周前
NO CONTEXT HUMANS
2周前
NO CONTEXT HUMANS
2周前
NO CONTEXT HUMANS
2周前
Noah Smith 🐇
2周前
NO CONTEXT HUMANS
2周前
挪威小林翠子🐈⬛
2周前
Rohan Paul
2周前
Your brain's next 5 seconds, predicted by AI Transformer predicts brain activity patterns 5 seconds into future using just 21 seconds of fMRI data Achieves 0.997 correlation using modified time-series Transformer architecture ----- 🧠 Original Problem: Predicting future brain states from fMRI data remains challenging, especially for patients who can't undergo long scanning sessions. Current methods require extensive scan times and lack accuracy in short-term predictions. ----- 🔬 Solution in this Paper: → The paper introduces a modified time series Transformer with 4 encoder and 4 decoder layers, each containing 8 attention heads → The model takes a 30-timepoint window covering 379 brain regions as input and predicts the next brain state → Training uses Human Connectome Project data from 1003 healthy adults, with preprocessing including spatial smoothing and bandpass filtering → Unlike traditional approaches, this model omits look-ahead masking, simplifying prediction for single future timepoints ----- 🎯 Key Insights: → Temporal dependencies in brain states can be effectively captured using self-attention mechanisms → Short input sequences (21.6s) suffice for accurate predictions → Error accumulation follows a Markov chain pattern in longer predictions → The model preserves functional connectivity patterns matching known brain organization ----- 📊 Results: → Single timepoint prediction achieves MSE of 0.0013 → Accurate predictions up to 5.04 seconds with correlation >0.85 → First 7 predicted timepoints maintain high accuracy → Outperforms BrainLM with 20-timepoint MSE of 0.26 vs 0.568
Wild Clips
2周前
non aesthetic things
3周前