Search Results for author: Fangzhen Lin

Found 9 papers, 3 papers with code

AMR-DA: Data Augmentation by Abstract Meaning Representation

1 code implementation Findings (ACL) 2022 Ziyi Shou, Yuxin Jiang, Fangzhen Lin

To evaluate the effectiveness of our method, we apply it to the tasks of semantic textual similarity (STS) and text classification.

Data Augmentation Semantic Textual Similarity +5

On Computing Universal Plans for Partially Observable Multi-Agent Path Finding

no code implementations25 May 2023 Fengming Zhu, Fangzhen Lin

Given an arbitrary two-dimensional map and a profile of goals for the agents, the system finds a feasible universal plan for each agent that ensures no collision with others.

Multi-Agent Path Finding

Using Language Models For Knowledge Acquisition in Natural Language Reasoning Problems

no code implementations4 Apr 2023 Fangzhen Lin, Ziyi Shou, Chengcai Chen

For a natural language problem that requires some non-trivial reasoning to solve, there are at least two ways to do it using a large language model (LLM).

Language Modelling Large Language Model

Backward Imitation and Forward Reinforcement Learning via Bi-directional Model Rollouts

no code implementations4 Aug 2022 Yuxin Pan, Fangzhen Lin

Traditional model-based reinforcement learning (RL) methods generate forward rollout traces using the learnt dynamics model to reduce interactions with the real environment.

Generative Adversarial Network Model-based Reinforcement Learning +2

PocketNN: Integer-only Training and Inference of Neural Networks via Direct Feedback Alignment and Pocket Activations in Pure C++

1 code implementation8 Jan 2022 Jaewoo Song, Fangzhen Lin

In this paper we propose PocketNN, a light and self-contained proof-of-concept framework in pure C++ for the training and inference of DNNs using only integers.


Computing Class Hierarchies from Classifiers

no code implementations2 Dec 2021 Kai Kang, Fangzhen Lin

In this paper, we propose a novel algorithm for automatically acquiring a class hierarchy from a classifier which is often a large neural network these days.

Incorporating EDS Graph for AMR Parsing

no code implementations Joint Conference on Lexical and Computational Semantics 2021 Ziyi Shou, Fangzhen Lin

AMR (Abstract Meaning Representation) and EDS (Elementary Dependency Structures) are two popular meaning representations in NLP/NLU.

AMR Parsing

Faster and Safer Training by Embedding High-Level Knowledge into Deep Reinforcement Learning

no code implementations22 Oct 2019 Haodi Zhang, Zihang Gao, Yi Zhou, Hao Zhang, Kaishun Wu, Fangzhen Lin

Deep reinforcement learning has been successfully used in many dynamic decision making domains, especially those with very large state spaces.

Decision Making reinforcement-learning +1

Cannot find the paper you are looking for? You can Submit a new open access paper.