Search Results for author: Shou-De Lin

Found 41 papers, 8 papers with code

Enhance the Robustness of Text-Centric Multimodal Alignments

no code implementations6 Jul 2024 Ting-Yu Yen, Yun-Da Tsai, Keng-Te Liao, Shou-De Lin

This study assesses the quality and robustness of multimodal representations in the presence of missing entries, noise, or absent modalities, revealing that current text-centric alignment methods compromise downstream robustness.

Transferable Embedding Inversion Attack: Uncovering Privacy Risks in Text Embeddings without Model Queries

1 code implementation12 Jun 2024 Yu-Hsiang Huang, YuChe Tsai, Hsiang Hsiao, Hong-Yi Lin, Shou-De Lin

This study investigates the privacy risks associated with text embeddings, focusing on the scenario where attackers cannot access the original embedding model.

LinearAPT: An Adaptive Algorithm for the Fixed-Budget Thresholding Linear Bandit Problem

no code implementations10 Mar 2024 Yun-Ang Wu, Yun-Da Tsai, Shou-De Lin

In this study, we delve into the Thresholding Linear Bandit (TLB) problem, a nuanced domain within stochastic Multi-Armed Bandit (MAB) problems, focusing on maximizing decision accuracy against a linearly defined threshold under resource constraints.

Computational Efficiency Decision Making

Text-centric Alignment for Multi-Modality Learning

no code implementations12 Feb 2024 Yun-Da Tsai, Ting-Yu Yen, Pei-Fu Guo, Zhe-Yan Li, Shou-De Lin

This research paper addresses the challenge of modality mismatch in multimodal learning, where the modalities available during inference differ from those available at training.

In-Context Learning

lil'HDoC: An Algorithm for Good Arm Identification under Small Threshold Gap

no code implementations29 Jan 2024 Tzu-Hsien Tsai, Yun-Da Tsai, Shou-De Lin

We demonstrate that the sample complexity of the first $\lambda$ output arm in lil'HDoC is bounded by the original HDoC algorithm, except for one negligible term, when the distance between the expected reward and threshold is small.

A Collaborative Filtering-Based Two Stage Model with Item Dependency for Course Recommendation

no code implementations1 Nov 2023 Eric L. Lee, Tsung-Ting Kuo, Shou-De Lin

We point out several challenges in applying the existing CF-models to build a course recommendation engine, including the lack of rating and meta-data, the imbalance of course registration distribution, and the demand of course dependency modeling.

Collaborative Filtering Recommendation Systems

PSGText: Stroke-Guided Scene Text Editing with PSP Module

no code implementations20 Oct 2023 Felix Liawi, Yun-Da Tsai, Guan-Lun Lu, Shou-De Lin

Initially, we introduce a text-swapping network that seamlessly substitutes the original text with the desired replacement.

Diversity Scene Text Editing

Towards Optimizing with Large Language Models

no code implementations8 Oct 2023 Pei-Fu Guo, Ying-Hsuan Chen, Yun-Da Tsai, Shou-De Lin

In this work, we conduct an assessment of the optimization capabilities of LLMs across various tasks and data sizes.

Environment Diversification with Multi-head Neural Network for Invariant Learning

no code implementations17 Aug 2023 Bo-Wei Huang, Keng-Te Liao, Chang-Sheng Kao, Shou-De Lin

On this issue, a research direction, invariant learning, has been proposed to extract invariant features insensitive to the distributional changes.

Differential Good Arm Identification

no code implementations13 Mar 2023 Yun-Da Tsai, Tzu-Hsien Tsai, Shou-De Lin

This paper targets a variant of the stochastic multi-armed bandit problem called good arm identification (GAI).

Toward More Generalized Malicious URL Detection Models

1 code implementation21 Feb 2022 YunDa Tsai, Cayon Liow, Yin Sheng Siang, Shou-De Lin

This paper reveals a data bias issue that can severely affect the performance while conducting a machine learning model for malicious URL detection.

BIG-bench Machine Learning Interpretable Machine Learning

Explainable and Sparse Representations of Academic Articles for Knowledge Exploration

no code implementations COLING 2020 Keng-Te Liao, Zhihong Shen, Chiyuan Huang, Chieh-Han Wu, PoChun Chen, Kuansan Wang, Shou-De Lin

Provided with the interpretable concepts and knowledge encoded in a pre-trained neural model, we investigate whether the tagged concepts can be applied to a broader class of applications.

Glyph2Vec: Learning Chinese Out-of-Vocabulary Word Embedding from Glyphs

no code implementations ACL 2020 Hong-You Chen, Sz-Han Yu, Shou-De Lin

Chinese NLP applications that rely on large text often contain huge amounts of vocabulary which are sparse in corpus.

Multiple Text Style Transfer by using Word-level Conditional Generative Adversarial Network with Two-Phase Training

no code implementations IJCNLP 2019 Chih-Te Lai, Yi-Te Hong, Hong-You Chen, Chi-Jen Lu, Shou-De Lin

The objective of non-parallel text style transfer, or controllable text generation, is to alter specific attributes (e. g. sentiment, mood, tense, politeness, etc) of a given text while preserving its remaining attributes and content.

Attribute Generative Adversarial Network +2

Controlling Sequence-to-Sequence Models - A Demonstration on Neural-based Acrostic Generator

no code implementations IJCNLP 2019 Liang-Hsin Shen, Pei-Lun Tai, Chao-Chung Wu, Shou-De Lin

An acrostic is a form of writing that the first token of each line (or other recurring features in the text) forms a meaningful sequence.

Text Generation

MetricGAN: Generative Adversarial Networks based Black-box Metric Scores Optimization for Speech Enhancement

5 code implementations13 May 2019 Szu-Wei Fu, Chien-Feng Liao, Yu Tsao, Shou-De Lin

Adversarial loss in a conditional generative adversarial network (GAN) is not designed to directly optimize evaluation metrics of a target task, and thus, may not always guide the generator in a GAN to generate data with improved metric scores.

Generative Adversarial Network Speech Enhancement


no code implementations ICLR 2019 Chih-Kuan Yeh, Ian E. H. Yen, Hong-You Chen, Chun-Pei Yang, Shou-De Lin, Pradeep Ravikumar

State-of-the-art deep neural networks (DNNs) typically have tens of millions of parameters, which might not fit into the upper levels of the memory hierarchy, thus increasing the inference time and energy consumption significantly, and prohibiting their use on edge devices such as mobile phones.

A Regulation Enforcement Solution for Multi-agent Reinforcement Learning

no code implementations29 Jan 2019 Fan-Yun Sun, Yen-Yu Chang, Yueh-Hua Wu, Shou-De Lin

If artificially intelligent (AI) agents make decisions on behalf of human beings, we would hope they can also follow established regulations while interacting with humans or other AI agents.

AI Agent Management +3

Attribute-aware Collaborative Filtering: Survey and Classification

no code implementations20 Oct 2018 Wen-Hao Chen, Chin-Chi Hsu, Yi-An Lai, Vincent Liu, Mi-Yen Yeh, Shou-De Lin

Attribute-aware CF models aims at rating prediction given not only the historical rating from users to items, but also the information associated with users (e. g. age), items (e. g. price), or even ratings (e. g. rating time).

Attribute Classification +2

ANS: Adaptive Network Scaling for Deep Rectifier Reinforcement Learning Models

no code implementations6 Sep 2018 Yueh-Hua Wu, Fan-Yun Sun, Yen-Yu Chang, Shou-De Lin

This work provides a thorough study on how reward scaling can affect performance of deep reinforcement learning agents.

reinforcement-learning Reinforcement Learning (RL)

A Memory-Network Based Solution for Multivariate Time-Series Forecasting

2 code implementations6 Sep 2018 Yen-Yu Chang, Fan-Yun Sun, Yueh-Hua Wu, Shou-De Lin

Inspired by Memory Network proposed for solving the question-answering task, we propose a deep learning based model named Memory Time-series network (MTNet) for time series forecasting.

Multivariate Time Series Forecasting Question Answering +1

A Low-Cost Ethics Shaping Approach for Designing Reinforcement Learning Agents

1 code implementation12 Dec 2017 Yueh-Hua Wu, Shou-De Lin

This paper proposes a low-cost, easily realizable strategy to equip a reinforcement learning (RL) agent the capability of behaving ethically.

Ethics reinforcement-learning +1

Towards a More Reliable Privacy-preserving Recommender System

no code implementations21 Nov 2017 Jia-Yun Jiang, Cheng-Te Li, Shou-De Lin

This paper proposes a privacy-preserving distributed recommendation framework, Secure Distributed Collaborative Filtering (SDCF), to preserve the privacy of value, model and existence altogether.

Collaborative Filtering Privacy Preserving +1

Latent Feature Lasso

no code implementations ICML 2017 Ian En-Hsu Yen, Wei-Cheng Lee, Sung-En Chang, Arun Sai Suggala, Shou-De Lin, Pradeep Ravikumar

The latent feature model (LFM), proposed in (Griffiths \& Ghahramani, 2005), but possibly with earlier origins, is a generalization of a mixture model, where each instance is generated not from a single latent class but from a combination of latent features.

Toward Implicit Sample Noise Modeling: Deviation-driven Matrix Factorization

no code implementations28 Oct 2016 Guang-He Lee, Shao-Wen Yang, Shou-De Lin

Specifically, by modeling and learning the deviation of data, we design a novel matrix factorization model.

A Dual Augmented Block Minimization Framework for Learning with Limited Memory

no code implementations NeurIPS 2015 Ian En-Hsu Yen, Shan-Wei Lin, Shou-De Lin

In past few years, several techniques have been proposed for training of linear Support Vector Machine (SVM) in limited-memory setting, where a dual block-coordinate descent (dual-BCD) method was used to balance cost spent on I/O and computation.

Sparse Random Feature Algorithm as Coordinate Descent in Hilbert Space

no code implementations NeurIPS 2014 Ian En-Hsu Yen, Ting-Wei Lin, Shou-De Lin, Pradeep K. Ravikumar, Inderjit S. Dhillon

In this paper, we propose a Sparse Random Feature algorithm, which learns a sparse non-linear predictor by minimizing an $\ell_1$-regularized objective function over the Hilbert Space induced from kernel function.

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