LambdaNetworks: Modeling Long-Range Interactions Without Attention
This work addresses the computational inefficiency of attention mechanisms in deep learning for tasks like image processing, offering a faster alternative with competitive performance.
The authors introduced LambdaNetworks, a framework using lambda layers to model long-range interactions without attention, which outperformed convolutional and attentional models on ImageNet classification, COCO object detection, and instance segmentation while being more computationally efficient, with LambdaResNets achieving up to 9.5x speed-up over EfficientNet checkpoints.
We present lambda layers -- an alternative framework to self-attention -- for capturing long-range interactions between an input and structured contextual information (e.g. a pixel surrounded by other pixels). Lambda layers capture such interactions by transforming available contexts into linear functions, termed lambdas, and applying these linear functions to each input separately. Similar to linear attention, lambda layers bypass expensive attention maps, but in contrast, they model both content and position-based interactions which enables their application to large structured inputs such as images. The resulting neural network architectures, LambdaNetworks, significantly outperform their convolutional and attentional counterparts on ImageNet classification, COCO object detection and COCO instance segmentation, while being more computationally efficient. Additionally, we design LambdaResNets, a family of hybrid architectures across different scales, that considerably improves the speed-accuracy tradeoff of image classification models. LambdaResNets reach excellent accuracies on ImageNet while being 3.2 - 4.4x faster than the popular EfficientNets on modern machine learning accelerators. When training with an additional 130M pseudo-labeled images, LambdaResNets achieve up to a 9.5x speed-up over the corresponding EfficientNet checkpoints.