Search Results for author:
Found 9 papers, 1 papers with code
RefConv: Re-parameterized Refocusing Convolution for Powerful ConvNets
We propose Re-parameterized Refocusing Convolution (RefConv) as a replacement for regular convolutional layers, which is a plug-and-play module to improve the performance without any inference costs.
X-MLP: A Patch Embedding-Free MLP Architecture for Vision
However, existing vision MLP architectures always depend on convolution for patch embedding.
Learn to Enhance the Negative Information in Convolutional Neural Network
In this way, LENI can enhance the model representational capacity significantly while maintaining the original advantages of ReLU.
FalconNet: Factorization for the Light-weight ConvNets
To address these issues, we factorize the four vital components of light-weight CNNs from coarse to fine and redesign them: i) we design a light-weight overall architecture termed LightNet, which obtains better performance by simply implementing the basic blocks of other light-weight CNNs; ii) we abstract a Meta Light Block, which consists of spatial operator and channel operator and uniformly describes current basic blocks; iii) we raise RepSO which constructs multiple spatial operator branches to enhance the representational ability; iv) we raise the concept of receptive range, guided by which we raise RefCO to sparsely factorize the channel operator.
Evolution: A Unified Formula for Feature Operators from a High-level Perspective
Traditionally, different types of feature operators (e. g., convolution, self-attention and involution) utilize different approaches to extract and aggregate the features.
SA-GD: Improved Gradient Descent Learning Strategy with Simulated Annealing
Compared to the baseline models with traditional gradient descent algorithm, models with SA-GD algorithm possess better generalization ability without sacrificing the efficiency and stability of model convergence.
Interflow: Aggregating Multi-layer Feature Mappings with Attention Mechanism
However, it can be difficulty to settle the optimal network depth and make the middle layers learn distinguished features.
Jitter: Random Jittering Loss Function
As Jitter point acting as a random factor, we actually add some randomness to the loss function, which is consistent with the fact that there exists innumerable random behaviors in the learning process of the machine learning model and is supposed to make the model more robust.
Reborn Mechanism: Rethinking the Negative Phase Information Flow in Convolutional Neural Network
This paper proposes a novel nonlinear activation mechanism typically for convolutional neural network (CNN), named as reborn mechanism.
