Search Results for author:
Found 100 papers, 21 papers with code
MetaWeighting: Learning to Weight Tasks in Multi-Task Learning
Task weighting, which assigns weights on the including tasks during training, significantly matters the performance of Multi-task Learning (MTL); thus, recently, there has been an explosive interest in it.
MedDialog: Large-scale Medical Dialogue Datasets
We also study the transferability of models trained on MedDialog to low-resource medical dialogue generation tasks.
Defense against Prompt Injection Attacks via Mixture of Encodings
Large Language Models (LLMs) have emerged as a dominant approach for a wide range of NLP tasks, with their access to external information further enhancing their capabilities.
Improving the Language Understanding Capabilities of Large Language Models Using Reinforcement Learning
In SFT, task-specific prompts are concatenated with input queries and ground-truth labels, optimizing with next-token prediction.
BiDoRA: Bi-level Optimization-Based Weight-Decomposed Low-Rank Adaptation
In BiDoRA, the direction and magnitude components are optimized on two distinct datasets at different optimization levels, mitigating the risk of overfitting.
Natural Language Understanding
parameter-efficient fine-tuning
+3
TapWeight: Reweighting Pretraining Objectives for Task-Adaptive Pretraining
Large-scale general domain pretraining followed by downstream-specific finetuning has become a predominant paradigm in machine learning.
Molecular Property Prediction
Natural Language Understanding
+1
AM-SAM: Automated Prompting and Mask Calibration for Segment Anything Model
Segment Anything Model (SAM) has gained significant recognition in the field of semantic segmentation due to its versatile capabilities and impressive performance.
LoGra-Med: Long Context Multi-Graph Alignment for Medical Vision-Language Model
State-of-the-art medical multi-modal large language models (med-MLLM), like LLaVA-Med or BioMedGPT, leverage instruction-following data in pre-training.
Generative AI Enables Medical Image Segmentation in Ultra Low-Data Regimes
While deep learning has excelled in automating this task, a major hurdle is the need for numerous annotated segmentation masks, which are resource-intensive to produce due to the required expertise and time.
Accelerating Transformers with Spectrum-Preserving Token Merging
Increasing the throughput of the Transformer architecture, a foundational component used in numerous state-of-the-art models for vision and language tasks (e. g., GPT, LLaVa), is an important problem in machine learning.
Generalizable and Stable Finetuning of Pretrained Language Models on Low-Resource Texts
Pretrained Language Models (PLMs) have advanced Natural Language Processing (NLP) tasks significantly, but finetuning PLMs on low-resource datasets poses significant challenges such as instability and overfitting.
BiLoRA: A Bi-level Optimization Framework for Overfitting-Resilient Low-Rank Adaptation of Large Pre-trained Models
Low-rank adaptation (LoRA) is a popular method for fine-tuning large-scale pre-trained models in downstream tasks by learning low-rank incremental matrices.
AutoLoRA: Automatically Tuning Matrix Ranks in Low-Rank Adaptation Based on Meta Learning
Large-scale pretraining followed by task-specific finetuning has achieved great success in various NLP tasks.
Downstream Task Guided Masking Learning in Masked Autoencoders Using Multi-Level Optimization
Masked Autoencoder (MAE) is a notable method for self-supervised pretraining in visual representation learning.
Token-Specific Watermarking with Enhanced Detectability and Semantic Coherence for Large Language Models
Large language models generate high-quality responses with potential misinformation, underscoring the need for regulation by distinguishing AI-generated and human-written texts.
BLO-SAM: Bi-level Optimization Based Overfitting-Preventing Finetuning of SAM
Secondly, SAM faces challenges in excelling at specific downstream tasks, like medical imaging, due to a disparity between the distribution of its pretraining data, which predominantly consists of general-domain images, and the data used in downstream tasks.
On the Out of Distribution Robustness of Foundation Models in Medical Image Segmentation
Constructing a robust model that can effectively generalize to test samples under distribution shifts remains a significant challenge in the field of medical imaging.
LVM-Med: Learning Large-Scale Self-Supervised Vision Models for Medical Imaging via Second-order Graph Matching
While pre-trained deep networks on ImageNet and vision-language foundation models trained on web-scale data are prevailing approaches, their effectiveness on medical tasks is limited due to the significant domain shift between natural and medical images.
DrugChat: Towards Enabling ChatGPT-Like Capabilities on Drug Molecule Graphs
The DrugChat system consists of a graph neural network (GNN), a large language model (LLM), and an adaptor.
Learning by Grouping: A Multilevel Optimization Framework for Improving Fairness in Classification without Losing Accuracy
Our proposed framework involves three stages of learning, which are formulated as a three-level optimization problem: (i) learning to group problems into different subgroups; (ii) learning group-specific sub-models for problem-solving; and (iii) updating group assignments of training examples by minimizing the validation loss.
A Comprehensive Survey on Pretrained Foundation Models: A History from BERT to ChatGPT
This study provides a comprehensive review of recent research advancements, challenges, and opportunities for PFMs in text, image, graph, as well as other data modalities.
Learning Compiler Pass Orders using Coreset and Normalized Value Prediction
Finding the optimal pass sequence of compilation can lead to a significant reduction in program size and/or improvement in program efficiency.
DRG-Net: Interactive Joint Learning of Multi-lesion Segmentation and Classification for Diabetic Retinopathy Grading
Diabetic Retinopathy (DR) is a leading cause of vision loss in the world, and early DR detection is necessary to prevent vision loss and support an appropriate treatment.
Joint Self-Supervised Image-Volume Representation Learning with Intra-Inter Contrastive Clustering
Recent breakthroughs in self-supervised learning (SSL) offer the ability to overcome the lack of labeled training samples by learning feature representations from unlabeled data.
Type Information Utilized Event Detection via Multi-Channel GNNs in Electrical Power Systems
To address the above issues, we propose a Multi-Channel graph neural network utilizing Type information for Event Detection in power systems, named MC-TED, leveraging a semantic channel and a topological channel to enrich information interaction from short texts.
Saliency-Aware Neural Architecture Search
At the second stage, saliency maps are generated using the trained model.
Betty: An Automatic Differentiation Library for Multilevel Optimization
Gradient-based multilevel optimization (MLO) has gained attention as a framework for studying numerous problems, ranging from hyperparameter optimization and meta-learning to neural architecture search and reinforcement learning.
Not All Patches are What You Need: Expediting Vision Transformers via Token Reorganizations
Second, by maintaining the same computational cost, our method empowers ViTs to take more image tokens as input for recognition accuracy improvement, where the image tokens are from higher resolution images.
Ranked #4 on
Efficient ViTs
on ImageNet-1K (with DeiT-S)
Self-directed Machine Learning
Inspired by the concept of self-directed human learning, we introduce the principal concept of Self-directed Machine Learning (SDML) and propose a framework for SDML.
Performance-Aware Mutual Knowledge Distillation for Improving Neural Architecture Search
In existing MKD methods, mutual knowledge distillation is performed between models without scrutiny: a worse-performing model is allowed to generate knowledge to train a better-performing model, which may lead to collective failures.
Learning from Mistakes based on Class Weighting with Application to Neural Architecture Search
In this work, we aim to investigate how effectively we can leverage this exceptional learning ability to improve machine learning models.
Improving Differentiable Architecture Search with a Generative Model
We introduce a training strategy called ``Differentiable Architecture Search with a Generative Model(DASGM)."
Learning from Mistakes -- A Framework for Neural Architecture Search
We propose a novel machine learning method called Learning From Mistakes (LFM), wherein the learner improves its ability to learn by focusing more on the mistakes during revision.
EViT: Expediting Vision Transformers via Token Reorganizations
Second, by maintaining the same computational cost, our method empowers ViTs to take more image tokens as input for recognition accuracy improvement, where the image tokens are from higher resolution images.
Learning by Examples Based on Multi-level Optimization
Learning by examples, which learns to solve a new problem by looking into how similar problems are solved, is an effective learning method in human learning.
On the Generation of Medical Dialogs for COVID-19
Training complex dialog generation models on small datasets bears high risk of overfitting.
Towards Visual Question Answering on Pathology Images
In this paper, we aim to develop a pathological visual question answering framework to analyze pathology images and answer medical questions related to these images.
Interleaving Learning, with Application to Neural Architecture Search
Inspired by the interleaving learning technique of humans, in this paper we explore whether this learning methodology is beneficial for improving the performance of machine learning models as well.
Learning by Teaching, with Application to Neural Architecture Search
In the LBT framework, a teacher model improves itself by teaching a student model to learn well.
Self-supervised Regularization for Text Classification
The SSL task is unsupervised, which is defined purely on input texts without using any human-provided labels.
Learning by Ignoring, with Application to Domain Adaptation
We propose a novel machine learning framework referred to as learning by ignoring (LBI).
Learning by Self-Explanation, with Application to Neural Architecture Search
In our approach, an explainer model improves its learning ability by trying to clearly explain to an audience model regarding how a prediction outcome is made.
Small-Group Learning, with Application to Neural Architecture Search
SGL is formulated as a multi-level optimization framework consisting of three learning stages: each learner trains a model independently and uses this model to perform pseudo-labeling; each learner trains another model using datasets pseudo-labeled by other learners; learners improve their architectures by minimizing validation losses.
DSRNA: Differentiable Search of Robust Neural Architectures
To address this problem, we propose methods to perform differentiable search of robust neural architectures.
Skillearn: Machine Learning Inspired by Humans' Learning Skills
To achieve this goal, we develop a general framework -- Skillearn, which provides a principled way to represent humans' learning skills mathematically and use the formally-represented skills to improve the training of ML models.
Learning by Passing Tests, with Application to Neural Architecture Search
We propose a multi-level optimization framework to formulate LPT, where the tester learns to create difficult and meaningful tests and the learner learns to pass these tests.
Validate and Enable Machine Learning in Industrial AI
However, manufacturers and solution partners need to understand how to implement and integrate an AI model into the existing industrial control system.
Discriminative Cross-Modal Data Augmentation for Medical Imaging Applications
While deep learning methods have shown great success in medical image analysis, they require a number of medical images to train.
Pathological Visual Question Answering
To deal with the issue that a publicly available pathology VQA dataset is lacking, we create PathVQA dataset.
TreeGAN: Incorporating Class Hierarchy into Image Generation
Based on these two ideas, we propose a TreeGAN model which consists of three modules: (1) a class hierarchy encoder (CHE) which takes the hierarchical structure of classes and their textual names as inputs and learns an embedding for each class; the embedding captures the hierarchical relationship among classes; (2) a conditional image generator (CIG) which takes the CHE-generated embedding of a class as input and generates a set of images belonging to this class; (3) a consistency checker which performs hierarchical classification on the generated images and checks whether the generated images are compatible with the class hierarchy; the consistency score is used to guide the CIG to generate hierarchy-compatible images.
Contrastive Self-supervised Learning for Graph Classification
A contrastive loss is defined to learn graph encoders by judging whether two augmented graphs are from the same original graph.
Transfer Learning or Self-supervised Learning? A Tale of Two Pretraining Paradigms
There has not been a clear understanding on what properties of data and tasks render one approach outperforms the other.
XRayGAN: Consistency-preserving Generation of X-ray Images from Radiology Reports
To address this problem, we develop methods to generate view-consistent, high-fidelity, and high-resolution X-ray images from radiology reports to facilitate radiology training of medical students.
Differentially-private Federated Neural Architecture Search
To address this problem, we propose federated neural architecture search (FNAS), where different parties collectively search for a differentiable architecture by exchanging gradients of architecture variables without exposing their data to other parties.
CERT: Contrastive Self-supervised Learning for Language Understanding
We evaluate CERT on 11 natural language understanding tasks in the GLUE benchmark where CERT outperforms BERT on 7 tasks, achieves the same performance as BERT on 2 tasks, and performs worse than BERT on 2 tasks.
On the Generation of Medical Dialogues for COVID-19
On these two datasets, we train several dialogue generation models based on Transformer, GPT, and BERT-GPT.
Sample-Efficient Deep Learning for COVID-19 Diagnosis Based on CT Scans
Besides, these works require a large number of CTs to train accurate diagnosis models, which are difficult to obtain.
MedDialog: Two Large-scale Medical Dialogue Datasets
Medical dialogue systems are promising in assisting in telemedicine to increase access to healthcare services, improve the quality of patient care, and reduce medical costs.
Identifying Radiological Findings Related to COVID-19 from Medical Literature
Coronavirus disease 2019 (COVID-19) has infected more than one million individuals all over the world and caused more than 55, 000 deaths, as of April 3 in 2020.
COVID-CT-Dataset: A CT Scan Dataset about COVID-19
Using this dataset, we develop diagnosis methods based on multi-task learning and self-supervised learning, that achieve an F1 of 0. 90, an AUC of 0. 98, and an accuracy of 0. 89.
PathVQA: 30000+ Questions for Medical Visual Question Answering
To achieve this goal, the first step is to create a visual question answering (VQA) dataset where the AI agent is presented with a pathology image together with a question and is asked to give the correct answer.
Specific and Shared Causal Relation Modeling and Mechanism-Based Clustering
The learned SSCM gives the specific causal knowledge for each individual as well as the general trend over the population.
Generalized Zero-shot ICD Coding
The International Classification of Diseases (ICD) is a list of classification codes for the diagnoses.
Explaining A Black-box By Using A Deep Variational Information Bottleneck Approach
Briefness and comprehensiveness are necessary in order to provide a large amount of information concisely when explaining a black-box decision system.
Adversarial Domain Adaptation Being Aware of Class Relationships
In this paper, we propose a novel relationship-aware adversarial domain adaptation (RADA) algorithm, which first utilizes a single multi-class domain discriminator to enforce the learning of inter-class dependency structure during domain-adversarial training and then aligns this structure with the inter-class dependencies that are characterized from training the label predictor on source domain.
Explaining a black-box using Deep Variational Information Bottleneck Approach
Briefness and comprehensiveness are necessary in order to provide a large amount of information concisely when explaining a black-box decision system.
Unsupervised Pseudo-Labeling for Extractive Summarization on Electronic Health Records
Extractive summarization is very useful for physicians to better manage and digest Electronic Health Records (EHRs).
Stackelberg GAN: Towards Provable Minimax Equilibrium via Multi-Generator Architectures
In this work, we give new results on the benefits of multi-generator architecture of GANs.
Multimodal Machine Learning for Automated ICD Coding
This study presents a multimodal machine learning model to predict ICD-10 diagnostic codes.
Missing Value Imputation Based on Deep Generative Models
In this paper, we propose a probabilistic framework based on deep generative models for MVI.
A Neural Architecture for Automated ICD Coding
The International Classification of Diseases (ICD) provides a hierarchy of diagnostic codes for classifying diseases.
Nonoverlap-Promoting Variable Selection
Variable selection is a classic problem in machine learning (ML), widely used to find important explanatory factors, and improve generalization performance and interpretability of ML models.
Orthogonality-Promoting Distance Metric Learning: Convex Relaxation and Theoretical Analysis
In this paper, we address these three issues by (1) seeking convex relaxations of the original nonconvex problems so that the global optimal is guaranteed to be achievable; (2) providing a formal analysis on OPR's capability of promoting balancedness; (3) providing a theoretical analysis that directly reveals the relationship between OPR and generalization performance.
Convolutional Neural Networks for Medical Diagnosis from Admission Notes
The hope is that the tool can be used to reduce mis-diagnosis.
Learning Less-Overlapping Representations
In representation learning (RL), how to make the learned representations easy to interpret and less overfitted to training data are two important but challenging issues.
Stacked Kernel Network
Similar to DNNs, a SKN is composed of multiple layers of hidden units, but each parameterized by a RKHS function rather than a finite-dimensional vector.
Diversity-Promoting Bayesian Learning of Latent Variable Models
We also extend our approach to "diversify" Bayesian nonparametric models where the number of components is infinite.
On the Automatic Generation of Medical Imaging Reports
To cope with these challenges, we (1) build a multi-task learning framework which jointly performs the pre- diction of tags and the generation of para- graphs, (2) propose a co-attention mechanism to localize regions containing abnormalities and generate narrations for them, (3) develop a hierarchical LSTM model to generate long paragraphs.
Effective Use of Bidirectional Language Modeling for Transfer Learning in Biomedical Named Entity Recognition
We also show that BiLM weight transfer leads to a faster model training and the pretrained model requires fewer training examples to achieve a particular F1 score.
Medical Diagnosis From Laboratory Tests by Combining Generative and Discriminative Learning
In hospital, physicians rely on massive clinical data to make diagnosis decisions, among which laboratory tests are one of the most important resources.
Towards Automated ICD Coding Using Deep Learning
Considering the complicated and dedicated process to assign correct codes to each patient admission based on overall diagnosis, we propose a hierarchical deep learning model with attention mechanism which can automatically assign ICD diagnostic codes given written diagnosis.
Predicting Discharge Medications at Admission Time Based on Deep Learning
Predicting discharge medications right after a patient being admitted is an important clinical decision, which provides physicians with guidance on what type of medication regimen to plan for and what possible changes on initial medication may occur during an inpatient stay.
Deep Determinantal Point Process for Large-Scale Multi-Label Classification
Experiments on the two datasets demonstrate the efficacy and efficiency of the proposed methods.
Uncorrelation and Evenness: a New Diversity-Promoting Regularizer
Latent space models (LSMs) provide a principled and effective way to extract hidden patterns from observed data.
Learning Latent Space Models with Angular Constraints
The large model capacity of latent space models (LSMs) enables them to achieve great performance on various applications, but meanwhile renders LSMs to be prone to overfitting.
A Constituent-Centric Neural Architecture for Reading Comprehension
Reading comprehension (RC), aiming to understand natural texts and answer questions therein, is a challenging task.
Poseidon: An Efficient Communication Architecture for Distributed Deep Learning on GPU Clusters
We show that Poseidon enables Caffe and TensorFlow to achieve 15. 5x speed-up on 16 single-GPU machines, even with limited bandwidth (10GbE) and the challenging VGG19-22K network for image classification.
Strategies and Principles of Distributed Machine Learning on Big Data
Taking the view that Big ML systems can benefit greatly from ML-rooted statistical and algorithmic insights --- and that ML researchers should therefore not shy away from such systems design --- we discuss a series of principles and strategies distilled from our recent efforts on industrial-scale ML solutions.
Latent Variable Modeling with Diversity-Inducing Mutual Angular Regularization
On two popular latent variable models --- restricted Boltzmann machine and distance metric learning, we demonstrate that MAR can effectively capture long-tail patterns, reduce model complexity without sacrificing expressivity and improve interpretability.
Poseidon: A System Architecture for Efficient GPU-based Deep Learning on Multiple Machines
To investigate how to adapt existing frameworks to efficiently support distributed GPUs, we propose Poseidon, a scalable system architecture for distributed inter-machine communication in existing DL frameworks.
Distributed Training of Deep Neural Networks with Theoretical Analysis: Under SSP Setting
We propose a distributed approach to train deep neural networks (DNNs), which has guaranteed convergence theoretically and great scalability empirically: close to 6 times faster on instance of ImageNet data set when run with 6 machines.
Distributed Machine Learning via Sufficient Factor Broadcasting
Matrix-parametrized models, including multiclass logistic regression and sparse coding, are used in machine learning (ML) applications ranging from computer vision to computational biology.
On the Generalization Error Bounds of Neural Networks under Diversity-Inducing Mutual Angular Regularization
Recently diversity-inducing regularization methods for latent variable models (LVMs), which encourage the components in LVMs to be diverse, have been studied to address several issues involved in latent variable modeling: (1) how to capture long-tail patterns underlying data; (2) how to reduce model complexity without sacrificing expressivity; (3) how to improve the interpretability of learned patterns.
Cauchy Principal Component Analysis
In this paper, we propose Cauchy Principal Component Analysis (Cauchy PCA), a very simple yet effective PCA method which is robust to various types of noise.
Large Scale Distributed Distance Metric Learning
In large scale machine learning and data mining problems with high feature dimensionality, the Euclidean distance between data points can be uninformative, and Distance Metric Learning (DML) is often desired to learn a proper similarity measure (using side information such as example data pairs being similar or dissimilar).
Crypto-Nets: Neural Networks over Encrypted Data
To achieve the privacy requirements, we use homomorphic encryption in the following protocol: the data owner encrypts the data and sends the ciphertexts to the third party to obtain a prediction from a trained model.
Distributed Machine Learning via Sufficient Factor Broadcasting
Matrix-parametrized models, including multiclass logistic regression and sparse coding, are used in machine learning (ML) applications ranging from computer vision to computational biology.
Petuum: A New Platform for Distributed Machine Learning on Big Data
What is a systematic way to efficiently apply a wide spectrum of advanced ML programs to industrial scale problems, using Big Models (up to 100s of billions of parameters) on Big Data (up to terabytes or petabytes)?
Integrating Document Clustering and Topic Modeling
Document clustering and topic modeling are two closely related tasks which can mutually benefit each other.
