IEEE

Dual Balanced Class-Incremental Learning With im-Softmax and Angular Rectification

Abstract: Owing to the superior performances, exemplar-based methods with knowledge distillation (KD) are widely applied in class incremental learning (CIL). However, it suffers from two drawbacks: 1) ...
Scientific Research Publishing

Multimodal Digital Phenotyping for Bipolar Disorder: Robust Mood-State Classification and Early Relapse Risk Monitoring ()

Bipolar Disorder, Digital Phenotyping, Multimodal Learning, Face/Voice/Phone, Mood Classification, Relapse Prediction, T-SNE, Ablation Share and Cite: de Filippis, R. and Al Foysal, A. (2025) ...
IEEE

SimA: Simple Softmax-free Attention for Vision Transformers

Abstract: Recently, vision transformers have become very popular. However, deploying them in many applications is computationally expensive partly due to the Softmax layer in the attention block. We ...
GitHub

Simple Softmax-free Attention for Vision Transformer (SimA)

@misc{https://doi.org/10.48550/arxiv.2206.08898, doi = {10.48550/ARXIV.2206.08898}, url = {https://arxiv.org/abs/2206.08898}, author = {Koohpayegani, Soroush Abbasi ...
GitHub

Beta Machine Learning Toolkit

Theoretical notes describing many of these algorithms are at the companion repository https://github.com/sylvaticus/MITx_6.86x. If you are looking for an introductory ...