More Like this

1. NeuroView-RNN: It’s About Time

   https://doi.org/10.1145/3531146.3533224  

   Barberan, Cj; Alemmohammad, Sina; Liu, Naiming; Balestriero, Randall; Baraniuk, Richard (June 2022, FAccT '22: 2022 ACM Conference on Fairness, Accountability, and Transparency)

   Recurrent Neural Networks (RNNs) are important tools for processing sequential data such as time-series or video. Interpretability is defined as the ability to be understood by a person and is different from explainability, which is the ability to be explained in a mathematical formulation. A key interpretability issue with RNNs is that it is not clear how each hidden state per time step contributes to the decision-making process in a quantitative manner. We propose NeuroView-RNN as a family of new RNN architectures that explains how all the time steps are used for the decision-making process. Each member of the family is derived from a standard RNN architecture by concatenation of the hidden steps into a global linear classifier. The global linear classifier has all the hidden states as the input, so the weights of the classifier have a linear mapping to the hidden states. Hence, from the weights, NeuroView-RNN can quantify how important each time step is to a particular decision. As a bonus, NeuroView-RNN also offers higher accuracy in many cases compared to the RNNs and their variants. We showcase the benefits of NeuroView-RNN by evaluating on a multitude of diverse time-series datasets. 

   more » « less
2. Low latency RNN inference with cellular batching

   https://doi.org/10.1145/3190508.3190541  

   Gao, Pin; Yu, Lingfan; Wu, Yongwei; Li, Jinyang (April 2018, EuroSys '18: Proceedings of the Thirteenth EuroSys Conference)

   Performing inference on pre-trained neural network models must meet the requirement of low-latency, which is often at odds with achieving high throughput. Existing deep learning systems use batching to improve throughput, which do not perform well when serving Recurrent Neural Networks with dynamic dataflow graphs. We propose the technique of cellular batching, which improves both the latency and throughput of RNN inference. Unlike existing systems that batch a fixed set of dataflow graphs, cellular batching makes batching decisions at the granularity of an RNN "cell" (a subgraph with shared weights) and dynamically assembles a batched cell for execution as requests join and leave the system. We implemented our approach in a system called BatchMaker. Experiments show that BatchMaker achieves much lower latency and also higher throughput than existing systems. 

   more » « less
3. Hierarchically-Attentive RNN for Album Summarization and Storytelling

   https://doi.org/10.18653/v1/D17-1101  

   Yu, Licheng; Bansal, Mohit; Berg, Tamara (January 2017, Empirical Methods in Natural Language Processing)

   We address the problem of end-to-end visual storytelling. Given a photo album, our model first selects the most representative (summary) photos, and then composes a natural language story for the album. For this task, we make use of the Visual Storytelling dataset and a model composed of three hierarchically-attentive Recurrent Neural Nets (RNNs) to: encode the album photos, select representative (summary) photos, and compose the story. Automatic and human evaluations show our model achieves better performance on selection, generation, and retrieval than baselines. 

   more » « less
4. Recurrent Hierarchical Topic-Guided RNN for Language Generation

   Guo, D.; Chen, B.; Lu, R.; Zhou, M. (July 2020, International Conference on Machine Learning)

   null (Ed.)
5. GRNN: Low-Latency and Scalable RNN Inference on GPUs

   https://doi.org/10.1145/3302424.3303949  

   Holmes, Connor; Mawhirter, Daniel; He, Yuxiong; Yan, Feng; Wu, Bo (January 2019, EuroSys'19)