DeepFix Sequence-to-Sequence Neural Network

A DeepFix Sequence-to-Sequence Neural Network is a Sequence-to-Sequence Neural Network with Attention Mechanism that can fix multiple errors in computer programs written in C Language.

• Context:
  • It was initially developed by Gupta et al. (2017).
• Example(s):
  • …
• Counter-Example(s):
  • DeepFix Convolutional Neural Network.
• See: DeepFix Program Repair System, Multi-Layered Neural Network, Sequence-to-Sequence Learning Task.

---

---

References

2017

• (Gupta et al., 2017) ⇒ Rahul Gupta, Soham Pal, Aditya Kanade, and Shirish Shevade. (2017). “DeepFix: Fixing Common C Language Errors by Deep Learning.” In: Proceedings of AAAI-2017.
  • QUOTE: We present an end-to-end solution, called DeepFix, that does not use any external tool to localize or fix errors. We use a compiler only to validate the fixes suggested by DeepFix. At the heart of DeepFix is a multi-layered sequence-to-sequence neural network with attention (Bahdanau, Cho and Bengio, 2014), comprising of an encoder recurrent neural network (RNN) to process the input and a decoder RNN with attention that generates the output. The network is trained to predict an erroneous program location along with the correct statement. DeepFix invokes it iteratively to fix multiple errors in the program one-by-one.
    

    Figure 2: The iterative repair strategy of DeepFix.

     DeepFix uses a simple yet effective iterative strategy to fix multiple errors in a program as shown in Figure 2 ... Read More >>.

---