Africa
A Unifying Post-Processing Framework for Multi-Objective Learn-to-Defer Problems
Learn-to-Defer is a paradigm that enables learning algorithms to work not in isolation but as a team with human experts.
NeuroBOL T: Resting-state EEG-to-fMRI Synthesis with Multi-dimensional Feature Mapping Yamin Li
Functional magnetic resonance imaging (fMRI) and electroencephalography (EEG) are the most commonly utilized non-invasive neuroimaging techniques, providing crucial insights into brain functionality.
Communication-Efficient Federated Group Distributionally Robust Optimization
In contrast to CV aR, KL is a softer regularizer to promote smoother and more stable learning.
Decision Mamba: A Multi-Grained State Space Model with Self-Evolution Regularization for Offline RL Qi Lv
Thus, this kind of fine-grained intrinsic connection among RSAs is intuitively beneficial for policy learning.
RCDN: Towards Robust Camera-Insensitivity Collaborative Perception via Dynamic Feature-based 3D Neural Modeling
Collaborative perception is dedicated to tackling the constraints of single-agent perception, such as occlusions, based on the multiple agents' multi-view sensor inputs. However, most existing works assume an ideal condition that all agents'
Kaleidoscope: Learnable Masks for Heterogeneous Multi-agent Reinforcement Learning
We further extend Kaleidoscope to critic ensembles in the context of actor-critic algorithms, which could help improve value estimations.
Easy Regional Contrastive Learning of Expressive Fashion Representations
However, inheriting such benefits of CLIP in the fashion domain is non-trivial in the presence of the notable domain gap.
MemoryFormer: Minimize Transformer Computation by Removing Fully-Connected Layers
However, the model size and corresponding computational complexity are constantly scaled up in pursuit of higher performance.