Softmax is $1/2$-Lipschitz: A tight bound across all $\ell_p$ norms
Pravin Nair
Action editor: Murat Erdogdu
https://openreview.net/forum?id=6dowaHsa6D
#softmax #lipschitz #norms
2
0
0
0
#Softmax
Latest posts tagged with #Softmax on Bluesky
Trending
#U.S. Foreign Policy
#F1
#Chinese Grand Prix
#SNL
#Venezuela Baseball
#AEW Collision
#Six Nations
#Newcastle United
#Arsenal vs Everton
#Trump and Epstein
#U.S. Foreign Policy
#F1
#Chinese Grand Prix
#SNL
#Venezuela Baseball
#AEW Collision
#Six Nations
#Newcastle United
#Arsenal vs Everton
#Trump and Epstein
Posts tagged #Softmax
File:The-Transformer-model-architecture.png Output Probabilities Softmax Linear Add&Norm Feed Forward Add&Norm Multi-Head Attention Add&Norm Masked Multi-Head Attention Positional Encoding Output Embedding ↑ Add&Norm Feed Forward Add&Norm Multi-Head Attention Positional Encoding Input Embedding ↑ Input Output (shifted right)
"The transformer approach it describes has become the main architecture of a wide variety of AI, such as #LargeLanguageModels"
#OutputProbabilities
#Softmax
Linear
#Add&Norm
#FeedForward
#MultiHead Attention
#MaskedMultiHead Attention
#PositionalEncoding
#OutputEmbedding
#FeedForward
0
0
0
0
Training More Robust Classification Model via Discriminative Loss and Gaussian Noise Injection
Hai-Vy Nguyen, Fabrice Gamboa, Sixin Zhang, Reda CHHAIBI, Serge Gratton, Thierry Giaccone
Action editor: Lei Feng
https://openreview.net/forum?id=RnLfJgvST2
#softmax #robust #classification
0
0
0
0
Softmax returns with No ID — a dark, fuzzy reflection on disconnection and the music industry’s demands. Chunky guitars and robotic grooves carry a refrain that lingers: “Nothing I do, will ever be good enough for you.”
🎧 Listen & read: www.blackplastic.co.uk/alternative-... #NewMusic #Softmax
1
1
0
0
Harmonic Loss Trains Interpretable AI Models
David D. Baek, Ziming Liu, Riya Tyagi, Max Tegmark
Action editor: Quanshi Zhang
https://openreview.net/forum?id=ZpSZ7pNoCs
#softmax #representations #trained
1
0
0
0
On the Expressiveness of Softmax Attention: A Recurrent Neural Network Perspective
New #TMLR-Paper-with-Video:
On the Expressiveness of Softmax Attention: A Recurrent Neural Network Perspective
Gabriel Mongaras, Eric C. Larson
https://tmlr.infinite-conf.org/paper_pages/PHcITOi3vV
#softmax #rnns #attention
1
0
0
0
On the Expressiveness of Softmax Attention: A Recurrent Neural Network Perspective
Gabriel Mongaras, Eric C. Larson
Action editor: Lingpeng Kong
https://openreview.net/forum?id=PHcITOi3vV
#softmax #rnns #attention
1
0
0
0
Is isotropy a good proxy for generalization in time series forecasting with transformers?
Rashed Shelim, Shengzhe Xu, Walid Saad, Naren Ramakrishnan
Action editor: Jacek Cyranka
https://openreview.net/forum?id=iUtDYVQzFq
#softmax #forecasting #embeddings
1
0
0
0
High-Dimensional Gaussian Process Regression with Soft Kernel Interpolation
Chris L Camaño, Daniel Huang
Action editor: Geoff Pleiss
https://openreview.net/forum?id=U9b2FIjvWU
#softmax #interpolation #interpolating
1
1
0
0
Probabilities of Chat LLMs Are Miscalibrated but Still Predict Correctness on Multiple-Choice Q&A
Benjamin Plaut, Khanh Xuan Nguyen, Tu Trinh
Action editor: Kamalika Chaudhuri
https://openreview.net/forum?id=E6LOh5vz5x
#softmax #accuracy #language
0
0
0
0
Toward Linearly Regularizing the Geometric Bottleneck of Linear Generalized Attention
Jiaxu Liu, Xinping Yi, Xiangyu Yin, Yuhang Song, Gaojie Jin, Xiaowei Huang
Action editor: Shuangfei Zhai
https://openreview.net/forum?id=Vpyg3fqXbl
#attention #softmax #memory
0
0
0
0
What is #softmax and why is it important for machine learning? Check out my refresher tutorial on multiclass classification in neural networks and how you can build your own from scratch in Snap!, (or) your favorite programming language:
snap.berkeley.edu/project?user...
1
1
2
1
TRA: Better Length Generalisation with Threshold Relative Attention
Mattia Opper, Roland Fernandez, Paul Smolensky, Jianfeng Gao
Action editor: Petar Veličković
https://openreview.net/forum?id=yNiBUc2hMW
#attention #softmax #tasks
1
0
0
0
Beyond Softmax: New Gradient Bandit Framework Expands Learning
Bandit framework swaps softmax’s independence for nested‑logit models, enabling correlated actions. Authors prove sublinear regret bounds, experiments matching softmax‑based methods. Read more: getnews.me/beyond-softmax-new-gradi... #gradientbandit #softmax
0
0
0
0
Catnat Function Introduced as Efficient Alternative to Softmax
Catnat provides a binary‑split alternative to softmax, yielding a diagonal Fisher information matrix; benchmarks in graph learning, VAEs and reinforcement learning report test performance. Read more: getnews.me/catnat-function-introduc... #catnat #softmax
0
0
0
0
Softmax Attention Outperforms Linear in Single-Location Regression
Softmax‑based attention achieves Bayes‑optimal error in single‑location regression, while linear attention cannot, and softmax consistently outperforms it in finite‑sample tests. getnews.me/softmax-attention-outper... #softmax #attention
0
0
0
0
#alife2025 #ALIFE #ISAL #AI #artificiallife #STEM #CrossLabs #CrossCompass #XC #AWS #softmax #koin #SakanaAI #Google #MITPress #cafedecolombia #wolfram #insilence #mlst #springer #scicomm #pci #meetkyoto #thedeck #hotelanteroomkyoto
0
0
0
0
The deadline for submission is September 29th, 2025.
#alife2025 #ISAL #AI #artificiallife #STEM #CrossLabs #CrossCompass #XC #AWS #softmax #koin #SakanaAI #Google #MITPress #cafedecolombia #wolfram #insilence #mlst #springer #scicomm #pci #meetkyoto #thedeck #hotelanteroomkyoto
0
0
0
0
Sign-up form: forms.gle/FDBwX9dcovT1...
#alife2025 #ALIFE #ISAL #AI #STEM #CrossLabs #CrossCompass #XC #AWS #softmax #koin #SakanaAI #Google #MITPress #cafedecolombia #wolfram #insilence #insilenceai #mlst #springer #scicomm #pci #meetkyoto #thedeck #hotelanteroomkyoto
3
0
0
0
New Similarity-Distance-Magnitude Activation Improves Model Robustness
Similarity-Distance-Magnitude (SDM) activation adds similarity and distance awareness to softmax, improving robustness to out-of-distribution inputs; submitted on 16 Sep 2025. getnews.me/new-similarity-distance-... #sdm #softmax #robustness
0
0
0
0
Reach out to us in terms of any query - 2025.alife.org
Thank you. ☺️
#alife2025 #ALIFE #ISAL #AI #artificiallife #STEM #CrossLabs #CrossCompass #XC #AWS #softmax #koin #SakanaAI #Google #MITPress #cafedecolombia #wolfram #insilence #insilenceai #mlst #springer #scicomm
2
0
0
0
#alife2025 #ALIFE #ISAL #AI #artificiallife #STEM #CrossLabs #CrossCompass #XC #AWS #softmax #koin #SakanaAI #Google #MITPress #cafedecolombia #wolfram #insilence #insilenceai #mlst #springer #scicomm
1
0
0
0
#alife2025 #ALIFE #ISAL #AI #artificiallife #STEM #CrossLabs #CrossCompass #XC #AWS #softmax #koin #SakanaAI #Google #MITPress #cafedecolombia #wolfram #insilence #insilenceai #mlst #springer #scicomm
1
0
0
0
#alife2025 #ALIFE #ISAL #AI #artificiallife #STEM #CrossLabs #CrossCompass #XC #AWS #softmax #koin #SakanaAI #Google #MITPress #cafedecolombia #wolfram #insilence #insilenceai #mlst #springer #scicomm
0
0
0
0
#alife2025 #ALIFE #ISAL #AI #artificiallife #STEM #CrossLabs #CrossCompass #XC #AWS #softmax #koin #SakanaAI #Google #MITPress #cafedecolombia #wolfram #insilence #insilenceai #mlst #springer #scicomm
1
0
0
0
#alife2025 #ALIFE #ISAL #AI #artificiallife #STEM #CrossLabs #CrossCompass #XC #AWS #softmax #koin #SakanaAI #Google #MITPress #wolfram #insilence #insilenceai #mlst #springer #scicomm
1
0
0
0
#alife2025 #ALIFE #ISAL #AI #artificiallife #STEM #CrossLabs #CrossCompass #XC #AWS #softmax #koin #SakanaAI #Google #MITPress #wolfram #insilence #insilenceai #mlst #springer #scicomm
1
0
0
0
#alife2025 #ALIFE #ISAL #AI #artificiallife #STEM #CrossLabs #CrossCompass #XC #AWS #softmax #koin #SakanaAI #Google #MITPress #wolfram #insilence #insilenceai #mlst #springer #scicomm
1
0
0
0