Efficient (Soft) Q-Learning for Text Generation with Limited Good Data
arXiv:2106.07704
Abstract
Maximum likelihood estimation (MLE) is the predominant algorithm for training text generation models. This paradigm relies on direct supervision examples, which is not applicable to many emerging applications, such as generating adversarial attacks or generating prompts to control language models. Reinforcement learning (RL) on the other hand offers a more flexible solution by allowing users to plug in arbitrary task metrics as reward. Yet previous RL algorithms for text generation, such as policy gradient (on-policy RL) and Q-learning (off-policy RL), are often notoriously inefficient or unstable to train due to the large sequence space and the sparse reward received only at the end of sequences. In this paper, we introduce a new RL formulation for text generation from the soft Q-learning (SQL) perspective. It enables us to draw from the latest RL advances, such as path consistency learning, to combine the best of on-/off-policy updates, and learn effectively from sparse reward. We apply the approach to a wide range of novel text generation tasks, including learning from noisy/negative examples, adversarial attacks, and prompt generation. Experiments show our approach consistently outperforms both task-specialized algorithms and the previous RL methods.
Code available at https://github.com/HanGuo97/soft-Q-learning-for-text-generation
References in corpus (10)
- Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation
- Prefix-Tuning: Optimizing Continuous Prompts for Generation
- An Actor-Critic Algorithm for Sequence Prediction
- AutoPrompt: Eliciting Knowledge from Language Models with Automatically Generated Prompts
- The GEM Benchmark: Natural Language Generation, its Evaluation and Metrics
- Unifying Human and Statistical Evaluation for Natural Language Generation
- End-to-End Offline Goal-Oriented Dialog Policy Learning via Policy Gradient
- Maximum Entropy RL (Provably) Solves Some Robust RL Problems
- Cold-Start Reinforcement Learning with Softmax Policy Gradient
- Batch Policy Gradient Methods for Improving Neural Conversation Models