More Like this

1. Learning to Allocate Time-Bound and Dynamic Tasks to Multiple Robots Using Covariant Attention Neural Networks

   https://doi.org/10.1115/1.4065883  

   Paul, Steve; Chowdhury, Souma (September 2024, Journal of Computing and Information Science in Engineering)

   Abstract In various applications of multi-robotics in disaster response, warehouse management, and manufacturing, tasks that are known a priori and tasks added during run time need to be assigned efficiently and without conflicts to robots in the team. This multi-robot task allocation (MRTA) process presents itself as a combinatorial optimization (CO) problem that is usually challenging to be solved in meaningful timescales using typical (mixed)integer (non)linear programming tools. Building on a growing body of work in using graph reinforcement learning to learn search heuristics for such complex CO problems, this paper presents a new graph neural network architecture called the covariant attention mechanism (CAM). CAM can not only generalize but also scale to larger problems than that encountered in training, and handle dynamic tasks. This architecture combines the concept of covariant compositional networks used here to embed the local structures in task graphs, with a context module that encodes the robots’ states. The encoded information is passed onto a decoder designed using multi-head attention mechanism. When applied to a class of MRTA problems with time deadlines, robot ferry range constraints, and multi-trip settings, CAM surpasses a state-of-the-art graph learning approach based on the attention mechanism, as well as a feasible random-walk baseline across various generalizability and scalability tests. Performance of CAM is also found to be at par with a high-performing non-learning baseline called BiG-MRTA, while noting up to a 70-fold improvement in decision-making efficiency over this baseline. 

   more » « less
2. G-CASCADE: Efficient Cascaded Graph Convolutional Decoding for 2D Medical Image Segmentation

   https://doi.org/10.1109/WACV57701.2024.00755  

   Rahman, Md Mostafijur; Marculescu, Radu (January 2024, IEEE)

   In this paper, we are the first to propose a new graph convolution-based decoder namely, Cascaded Graph Convolutional Attention Decoder (G-CASCADE), for 2D medical image segmentation. G-CASCADE progressively refines multi-stage feature maps generated by hierarchical transformer encoders with an efficient graph convolution block. The encoder utilizes the self-attention mechanism to capture long-range dependencies, while the decoder refines the feature maps preserving long-range information due to the global receptive fields of the graph convolution block. Rigorous evaluations of our decoder with multiple transformer encoders on five medical image segmentation tasks (i.e., Abdomen organs, Cardiac organs, Polyp lesions, Skin lesions, and Retinal vessels) show that our model outperforms other state-of-the-art (SOTA) methods. We also demonstrate that our decoder achieves better DICE scores than the SOTA CASCADE decoder with 80.8% fewer parameters and 82.3% fewer FLOPs. Our decoder can easily be used with other hierarchical encoders for general-purpose semantic and medical image segmentation tasks. 

   more » « less
3. Deep Reinforcement Learning for Sequence-to-Sequence Models

   https://doi.org/10.1109/TNNLS.2019.2929141  

   Keneshloo, Yaser; Shi, Tian; Ramakrishnan, Naren; Reddy, Chandan K. (July 2020, IEEE Transactions on Neural Networks and Learning Systems)

   In recent times, sequence-to-sequence (seq2seq) models have gained a lot of popularity and provide stateof-the-art performance in a wide variety of tasks, such as machine translation, headline generation, text summarization, speech-to-text conversion, and image caption generation. The underlying framework for all these models is usually a deep neural network comprising an encoder and a decoder. Although simple encoder–decoder models produce competitive results, many researchers have proposed additional improvements over these seq2seq models, e.g., using an attention-based model over the input, pointer-generation models, and self-attention models. However, such seq2seq models suffer from two common problems: 1) exposure bias and 2) inconsistency between train/test measurement. Recently, a completely novel point of view has emerged in addressing these two problems in seq2seq models, leveraging methods from reinforcement learning (RL). In this survey, we consider seq2seq problems from the RL point of view and provide a formulation combining the power of RL methods in decision-making with seq2seq models that enable remembering long-term memories. We present some of the most recent frameworks that combine the concepts from RL and deep neural networks. Our work aims to provide insights into some of the problems that inherently arise with current approaches and how we can address them with better RL models. We also provide the source code for implementing most of the RL models discussed in this paper to support the complex task of abstractive text summarization and provide some targeted experiments for these RL models, both in terms of performance and training time. 

   more » « less
4. Role Equivalence Attention for Label Propagation in Graph Neural Networks

   https://doi.org/10.1007/978-3-030-47436-2_42  

   Park, Hogun; Neville, Jennifer (April 2021, Lecture notes in computer science)

   Semi-supervised relational learning methods aim to classify nodes in a partially-labeled graph. While popular, existing methods using Graph Neural Networks (GNN) for semi-supervised relational learning have mainly focused on learning node representations by aggregating nearby attributes, and it is still challenging to leverage inferences about unlabeled nodes with few attributes—particularly when trying to exploit higher-order relationships in the network efficiently. To address this, we propose a Graph Neural Network architecture that incorporates patterns among the available class labels and uses (1) a Role Equivalence attention mechanism and (2) a mini-batch importance sampling method to improve efficiency when learning over high-order paths. In particular, our Role Equivalence attention mechanism is able to use nodes’ roles to learn how to focus on relevant distant neighbors, in order to adaptively reduce the increased noise that occurs when higher-order structures are considered. In experiments on six different real-world datasets, we show that our model (REGNN) achieves significant performance gains compared to other recent state-of-the-art baselines, particularly when higher-order paths are considered in the models. 

   more » « less
5. Parallel Scale-wise Attention Network for Effective Scene Text Recognition

   U. Sajid, M. Chow (July 2021, International Joint Conference on Neural Networks (IJCNN),)

   null (Ed.)

   The paper proposes a new text recognition network for scene-text images. Many state-of-the-art methods employ the attention mechanism either in the text encoder or decoder for the text alignment. Although the encoder-based attention yields promising results, these schemes inherit noticeable limitations. They perform the feature extraction (FE) and visual attention (VA) sequentially, which bounds the attention mechanism to rely only on the FE final single-scale output. Moreover, the utilization of the attention process is limited by only applying it directly to the single scale feature-maps. To address these issues, we propose a new multi-scale and encoder-based attention network for text recognition that performs the multi-scale FE and VA in parallel. The multi-scale channels also undergo regular fusion with each other to develop the coordinated knowledge together. Quantitative evaluation and robustness analysis on the standard benchmarks demonstrate that the proposed network outperforms the state-of-the-art in most cases. 

   more » « less