chho33 / LAMOL

Code for LAMOL: LAnguage MOdeling for Lifelong Language Learning

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

LAnguage-MOdeling-for-Lifelong-Language-Learning

Most research on lifelong learning (LLL) applies to images or games, but not language. We present LAMOL, a simple yet effective method for LLL based on language modeling. LAMOL replays pseudo-samples of previous tasks while requiring no extra memory or model capacity. Specifically, LAMOL is a language model that simultaneously learns to solve the task and generate training samples. When the model is trained for a new task, it generates pseudo-samples of previous tasks for training alongside data for the new task. The results show that LAMOL prevents catastrophic forgetting without any sign of intransigence and can perform up to five very different language tasks sequentially with only one model. Overall, LAMOL outperforms previous methods by a considerable margin and is only 2--3% worse than multitasking, which is usually considered the LLL upper bound.

Dataset

Task Dataset (Original Data Link)
Question Answering SQuAD version 1.1
Machine Translation IWSLT
Summarization CNN/DM
Natural Language Inference CNN/DM
Sentiment Analysis SST
Semantic Role Labeling QA‑SRL
Zero-Shot Relation Extraction QA‑ZRE
Goal-Oriented Dialogue WOZ
Semantic Parsing WikiSQL
Commonsense Reasoning MWSC
Text Classification AGNews, Yelp, Amazon, DBPedia, Yahoo

In order to unify the format of all the dataset, we first ran the code in https://github.com/salesforce/decaNLP to get the first 10 tranformed dataset, and then converted them into Squad-like format. For the last 5 dataset, we converted them directly. All converted dataset are available here.

Dependencies

  • Ubuntu >= 16.04
  • This code only supports the following GPUs:
    • NVIDIA Geforce RTX 2080TI
    • NVIDIA TESLA V100
  • python3
  • cuda 10.1
  • python packages are listed in requirements.txt

Setup

  1. Create the following two directories in wherever you want. (you can name the directories arbitrarily):
    • data directory: Where the dataset will be load by the model.
    • model directory: The place for the model to dump its outputs.
  2. Download the dataset: Download here and decompress it. After decompression, move all the files in the decompressed directory into data directory.
  3. Make a copy of env.example and save it as env. In env, set the value of DATA_DIR as data directory and set the value of MODEL_ROOT_DIR as model directory.

Training and Testing

train.sh and test.sh are the entrance for training and testing. Main options for them include:

Options Description
seq_train_type The mode to deal with a sequence of tasks. Mode include: lll|finetune|multitask|mas|ewc|gem. "lll" is the default value corresponding our proposed method. The others are the methods for comparing with our proposal.
tasks A sequence of tasks we want to train by seq_train_type. Leave a space between tasks after the --tasks tag. Tasks are the keys in TASK_DICT variable in settings.py
model_name The language model we want to use. The default is gpt2. Options include gpt2|openai-gpt,
gen_lm_sample_percentage This tag only works with --seq_train_type lll. The percentage of the size of the dataset will be generated as pseudo samples for our proposed method.
lm_lambda Lambda value for the loss function.
max_n_epochs Maximum epoch value for all tasks.
min_batch_size Minimum batch size for all tasks.
min_n_steps Minimum step for optimizing the model for all tasks.
n_train_epochs Epochs for training for all tasks.
n_gpu Number of gpu to be used.
reg_lambda Lambda value for mas and ewc.
top_k_lm Top k sampling for the language model.
top_k_qa Top k sampling for the qa model.
train_batch_size Batch size for all tasks. The default is 0. Once the value equals to 0, The batch size will be decided dynamically based on the memory usage of the gpu.

Training

Example:

If you want to train sst, srl and woz.en sequentially by our proposed method, run:

./train.sh --seq_train_type lll --tasks sst srl woz.en

Outputs:

If assigning multitask to --seq_train_type tag, the model will be dumped in $MODEL_ROOT_DIR / model_name / seq_train_type /TASK1_TASK2_... directory. Otherwise, it will be in $MODEL_ROOT_DIR / model_name / seq_train_type / TASK1_TASK2_... / TASK1, $MODEL_ROOT_DIR / model_name / seq_train_type / TASK1_TASK2_... / TASK2, ... directories.

For example, if you run:

./train.sh --seq_train_type multitask --tasks sst srl woz.en

Then the model will be dumped in: $MODEL_ROOT_DIR/gpt2/multitask/squad1_wikisql_sst_srl_woz.en.

If you run:

./train.sh --seq_train_type lll --model_name openai-gpt --gen_lm_sample_percentage 0.2 --tasks sst srl woz.en

Then the models will be dumped in the following directories: $MODEL_ROOT_DIR/openai-gpt/lll/sst_srl_woz.en_0.2/sst, $MODEL_ROOT_DIR/openai-gpt/lll/sst_srl_woz.en_0.2/srl, $MODEL_ROOT_DIR/openai-gpt/lll/sst_srl_woz.en_0.2/woz.en.

Testing

Example:

This example test the model trained on sst, srl and woz.en by finetune method.

./test.sh --seq_train_type finetune --tasks sst srl woz.en

Outputs:

After running testing program, the metrics: metrics.json will be dumped in the same directory of Training's outputs.

Acknowledgements:

  • We use the language model offered by transformers, a state-of-the-art natural language processing models library by Thomas Wolf et al.
  • The implementation of MAS follows MAS-Memory-Aware-Synapses, the Memory Aware Synapses method implementation code by Aljundi R. et al.
  • The implementation of GEM follows GradientEpisodicMemory, the Gradient Episodic Memory method implementation code by Lopez-Paz, David et al.
  • The implementation of fp16 (fp16.py, fp16util.py) is from Megatron-LM, the ongoing research training transformer language models at scale by NVIDIA.
  • Data format conversion refer to decaNLP, the Natural Language Decathlon: Multitask Learning as Question Answering implementation code by Bryan McCann et al.

About

Code for LAMOL: LAnguage MOdeling for Lifelong Language Learning

License:MIT License


Languages

Language:Python 99.8%Language:Shell 0.2%