Little research explores the correlation between the expressive ability and generalization ability of the low-rank adaptation (LoRA). Sharpness-Aware Minimization (SAM) improves model generalization for both Convolutional Neural Networks (CNNs) and Transformers by encouraging convergence to locally flat minima. However, the connection between sharpness and generalization has not been fully explored for LoRA due to the lack of tools to either empirically seek flat minima or develop theoretical methods. In this work, we propose Flat Minima LoRA (FMLoRA) and its efficient version i.e., EFMLoRA, to seek flat minima for LoRA. Concretely, we theoretically demonstrate that perturbations in the full parameter space can be transferred to the low-rank subspace. This approach eliminates the potential interference introduced by perturbations across multiple matrices in the low-rank subspace. Our extensive experiments on large language models and vision-language models demonstrate that EFMLoRA achieves optimization efficiency comparable to that of LoRA while simultaneously attaining comparable or even better performance. For example, on the GLUE dataset with RoBERTa-large, EFMLoRA outperforms LoRA and full fine-tuning by 1.0% and 0.5% on average, respectively. On vision-language models e.g., Qwen-VL-Chat, there are performance improvements of 1.5% and 1.0% on the SQA and VizWiz datasets, respectively. These empirical results also verify that the generalization of LoRA is closely related to sharpness, which is omitted by previous methods.

In recent years, denoising methods based on deep learning have achieved unparalleled performance at the cost of large computational complexity. In this work, we propose an Efficient Multi-stage Video Denoising algorithm, called EMVD, to drastically reduce the complexity while maintaining or even improving the performance. First, a fusion stage reduces the noise through a recursive combination of all past frames in the video. Then, a denoising stage removes the noise in the fused frame. Finally, a refinement stage restores the missing high frequency in the denoised frame. All stages operate on a transform-domain representation obtained by learnable and invertible linear operators which simultaneously increase accuracy and decrease complexity of the model. A single loss on the final output is sufficient for successful convergence, hence making EMVD easy to train. Experiments on real raw data demonstrate that EMVD outperforms the state of the art when complexity is constrained, and even remains competitive against methods whose complexities are several orders of magnitude higher. Further, the low complexity and memory requirements of EMVD enable real-time video denoising on commercial SoC in mobile devices.

Generative Adversarial Networks (GAN) have shown promising performance in image synthesis and unsupervised learning (USL). In most cases, however, the representations extracted from unsupervised GAN are usually unsatisfactory in other computer vision tasks. By using conditional GAN (CGAN), this problem could be solved to some extent, but the main drawback of such models is the necessity for labeled data. To improve both image synthesis quality and representation learning performance under the unsupervised setting, in this paper, we propose a simple yet effective Transformation Generative Adversarial Networks (TrGAN). In our approach, instead of capturing the joint distribution of image-label pairs p(x,y) as in conditional GAN, we try to estimate the joint distribution of transformed image t(x) and transformation t. Specifically, given a randomly sampled transformation t, we train the discriminator to give an estimate of input transformation, while following the adversarial training scheme of the original GAN. In addition, intermediate feature matching as well as feature-transform matching methods are introduced to strengthen the regularization on the generated features. To evaluate the quality of both generated samples and extracted representations, extensive experiments are conducted on four public datasets. The experimental results on the quality of both the synthesized images and the extracted representations demonstrate the effectiveness of our method.

With the development of deep neural networks, recent years have witnessed the increasing research interest on generative models. Specificly, Variational Auto-Encoders (VAE) and Generative Adversarial Networks (GAN) have achieved impressive results in various generative tasks. VAE is well established and theoretically elegant, but tends to generate blurry samples. In contrast, GAN has shown the advantage in visual quality of generated images, but suffers the difficulty in translating a random vector into a desired high-dimensional sample. As a result, the training dynamics in GAN are often unstable and the generated samples could collapse to limited modes. In this paper, we propose a new Auto-Encoder Generative Adversarial Networks (AEGAN), which takes advantages of both VAE and GAN. In our approach, instead of matching the encoded distribution of training samples to the prior Pz as in VAE, we map the random vector into the encoded latent space by adversarial training based on GAN. Besides, we also match the decoded distribution of training samples with that from random vectors. To evaluate our approach, we make comparison with other encoder-decoder based generative models on three public datasets. The experiments with both qualitative and quantitative results demonstrate the superiority of our algorithm over the comparison generative models.