publications

2024

1. NeurIPS

   Enhancing Feature Diversity Boosts Channel-Adaptive Vision Transformers

   Chau Pham, and Bryan A. Plummer

   In Advances in Neural Information Processing Systems (NeurIPS) 2024

   Abstract PDF Bibtex Code

   Multi-Channel Imaging (MCI) contains an array of challenges for encoding useful feature representations not present in traditional images. For example, images from two different satellites may both contain RGB channels, but the remaining channels can be different for each imaging source. Thus, MCI models must support a variety of channel configurations at test time. Recent work has extended traditional visual encoders for MCI, such as Vision Transformers (ViT), by supplementing pixel information with an encoding representing the channel configuration. However, these methods treat each channel equally, i.e., they do not consider the unique properties of each channel type, which can result in needless and potentially harmful redundancies in the learned features. For example, if RGB channels are always present, the other channels can focus on extracting information that cannot be captured by the RGB channels. To this end, we propose DiChaViT, which aims to enhance the diversity in the learned features of MCI-ViT models. This is achieved through a novel channel sampling strategy that encourages the selection of more distinct channel sets for training. Additionally, we employ regularization and initialization techniques to increase the likelihood that new information is learned from each channel. Many of our improvements are architecture agnostic and can be incorporated into new architectures as they are developed. Experiments on both satellite and cell microscopy datasets, CHAMMI, JUMP-CP, and So2Sat, report DiChaViT yields a 1.5 - 5.0% gain over the state-of-the-art. Our code is publicly available at https://github.com/chaudatascience/diverse_channel_vit.

   @InProceedings{phamDiChaVit2024,
   author = {Chau Pham and Bryan A. Plummer},
   title = {Enhancing Feature Diversity Boosts Channel-Adaptive Vision Transformers},
   booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
   year = {2024}}
2. ICLR

   Let Models Speak Ciphers: Multiagent Debate through Embeddings

   Chau Pham*, Boyi Liu*, Yingxiang Yang, Zhengyu Chen, Tianyi Liu, Jianbo Yuan, Bryan A. Plummer, Zhaoran Wang, and Hongxia Yang

   In International Conference on Learning Representations (ICLR) 2024

   Abstract PDF Bibtex Code

   Discussion and debate among Large Language Models (LLMs) have gained considerable attention due to their potential to enhance the reasoning ability of LLMs. Although natural language is an obvious choice for communication due to LLM’s language understanding capability, the token sampling step needed when generating natural language poses a potential risk of information loss, as it uses only one token to represent the model’s belief across the entire vocabulary. In this paper, we introduce a communication regime named CIPHER (Communicative Inter-Model Protocol Through Embedding Representation) to address this issue. Specifically, we remove the token sampling step from LLMs and let them communicate their beliefs across the vocabulary through the expectation of the raw transformer output embeddings. Remarkably, by deviating from natural language, CIPHER offers an advantage of encoding a broader spectrum of information without any modification to the model weights, outperforming the state-of-the-art LLM debate methods using natural language by 0.5-5.0% across five reasoning tasks and multiple open-source LLMs of varying sizes. This showcases the superiority and robustness of embeddings as an alternative "language" for communication among LLMs. We anticipate that CIPHER will inspire further exploration for the design of interactions within LLM agent systems, offering a new direction that could significantly influence future developments in the field.

   @InProceedings{phamCIPHER2024,
   author={Chau Pham and Boyi Liu and Yingxiang Yang and Zhengyu Chen and Tianyi Liu and Jianbo Yuan and Bryan A. Plummer and Zhaoran Wang and Hongxia Yang},
   title={Let Models Speak Ciphers: Multiagent Debate through Embeddings},
   booktitle={International Conference on Learning Representations (ICLR)},
   year={2024}}
3. WACV

   MixtureGrowth: Growing Neural Networks by Recombining Learned Parameters

   Chau Pham*, Piotr Teterwak*, Soren Nelson*, and Bryan A. Plummer

   In IEEE Winter Conference on Applications of Computer Vision (WACV) 2024

   Abstract PDF Bibtex Code

   Most deep neural networks are trained under fixed network architectures and require retraining when the architecture changes. If expanding the network’s size is needed, it is necessary to retrain from scratch, which is expensive. To avoid this, one can grow from a small network by adding random weights over time to gradually achieve the target network size. However, this naive approach falls short in practice as it brings too much noise to the growing process. Prior work tackled this issue by leveraging the already learned weights and training data for generating new weights through conducting a computationally expensive analysis step. In this paper, we introduce MixtureGrowth, a new approach to growing networks that circumvents the initialization overhead in prior work. Before growing, each layer in our model is generated with a linear combination of parameter templates. Newly grown layer weights are generated by using a new linear combination of existing templates for a layer. On one hand, these templates are already trained for the task, providing a strong initialization. On the other, the new coefficients provide flexibility for the added layer weights to learn something new. We show that our approach boosts top-1 accuracy over the state-of-the-art by 2-2.5% on CIFAR-100 and ImageNet datasets, while achieving comparable performance with fewer FLOPs to a larger network trained from scratch. Code is available at https://github.com/chaudatascience/mixturegrowth.

   @InProceedings{phamMixtureGrowth2024,
   author={Chau Pham and Piotr Teterwak and Soren Nelson and Bryan A. Plummer},
   title={MixtureGrowth: Growing Neural Networks by Recombining Learned Parameters},
   booktitle={IEEE Winter Conference on Applications of Computer Vision (WACV)},
   year={2024}}
4. IEEE BigData Cup

   Leveraging Large Language Models for Suicide Detection on Social Media with Limited Labels

   Vy Nguyen, and Chau Pham

   In IEEE International Conference on Big Data (Big Data) 2024

   Abstract PDF Bibtex Code

   The increasing frequency of suicidal thoughts highlights the importance of early detection and intervention. Social media platforms, where users often share personal experiences and seek help, could be utilized to identify individuals at risk. However, the large volume of daily posts makes manual review impractical. This paper explores the use of Large Language Models (LLMs) to automatically detect suicidal content in text-based social media posts. We propose a novel method for generating pseudo-labels for unlabeled data by prompting LLMs, along with traditional classification fine-tuning techniques to enhance label accuracy. To create a strong suicide detection model, we develop an ensemble approach involving prompting with Qwen2-72B-Instruct, and using fine-tuned models such as Llama3-8B, Llama3.1-8B, and Gemma2-9B. We evaluate our approach on the dataset of the Suicide Ideation Detection on Social Media Challenge, a track of the IEEE Big Data 2024 Big Data Cup. Additionally, we conduct a comprehensive analysis to assess the impact of different models and fine-tuning strategies on detection performance. Experimental results show that the ensemble model significantly improves the detection accuracy, by 5% points compared with the individual models. It achieves a weight F1 score of 0.770 on the public test set, and 0.731 on the private test set, providing a promising solution for identifying suicidal content in social media. Our analysis shows that the choice of LLMs affects the prompting performance, with larger models providing better accuracy. Our code and checkpoints are publicly available at https://github.com/khanhvynguyen/Suicide_Detection_LLMs.

   @InProceedings{nguyenLeveragingLLM2024,
   title={Leveraging Large Language Models for Suicide Detection on Social Media with Limited Labels},
   author={Vy Nguyen and Chau Pham},
   booktitle={2024 IEEE International Conference on Big Data (Big Data)},
   year={2024},
   keywords={large language models; text classification; limited labels; prompt engineering; suicide detection; social media analysis}}

2023

1. A Unified Framework for Connecting Noise Modeling to Boost Noise Detection

   Siqi Wang, Chau Pham, and Bryan A. Plummer

   Preprint (2023)

   Abstract PDF Bibtex Code

   Noisy labels can impair model performance, making the study of learning with noisy labels an important topic. Two conventional approaches are noise modeling and noise detection. However, these two methods are typically studied independently, and there has been limited work on their collaboration. In this work, we explore the integration of these two approaches, proposing an interconnected structure with three crucial blocks: noise modeling, source knowledge identification, and enhanced noise detection using noise source-knowledge-integration methods. This collaboration structure offers advantages such as discriminating hard negatives and preserving genuinely clean labels that might be suspiciously noisy. Our experiments on four datasets, featuring three types of noise and different combinations of each block, demonstrate the efficacy of these components’ collaboration. Our collaborative structure methods achieve up to a 10% increase in top-1 classification accuracy in synthesized noise datasets and 3-5% in real-world noisy datasets. The results also suggest that these components make distinct contributions to overall performance across various noise scenarios. These findings provide valuable insights for designing noisy label learning methods customized for specific noise scenarios in the future. Our code is accessible to the public.

   @article{wang2023unified,
   title={A Unified Framework for Connecting Noise Modeling to Boost Noise Detection},
   author={Siqi Wang and Chau Pham and Bryan A. Plummer},
   journal={arXiv preprint arXiv:2312.00827},
   year={2023}}
2. NeurIPS

   CHAMMI: A benchmark for channel-adaptive models in microscopy imaging

   Zitong Chen*, Chau Pham*, Siqi Wang, Michael Doron, Nikita Moshkov, Bryan A. Plummer, and Juan C. Caicedo

   In Advances in Neural Information Processing Systems (NeurIPS) 2023

   Abstract PDF Bibtex Code Data

   Most neural networks assume that input images have a fixed number of channels (three for RGB images). However, there are many settings where the number of channels may vary, such as microscopy images where the number of channels changes depending on instruments and experimental goals. Yet, there has not been a systemic attempt to create and evaluate neural networks that are invariant to the number and type of channels. As a result, trained models remain specific to individual studies and are hardly reusable for other microscopy settings. In this paper, we present a benchmark for investigating channel-adaptive models in microscopy imaging, which consists of 1) a dataset of varied-channel single-cell images, and 2) a biologically relevant evaluation framework. In addition, we adapted several existing techniques to create channel-adaptive models and compared their performance on this benchmark to fixed-channel, baseline models. We find that channel-adaptive models can generalize better to out-of-domain tasks and can be computationally efficient. We contribute a curated dataset and an evaluation API to facilitate objective comparisons in future research and applications.

   @InProceedings{ChenCHAMMI2023,
   author={Zitong Chen and Chau Pham and Siqi Wang and Michael Doron and Nikita Moshkov and Bryan A. Plummer and Juan C Caicedo},
   title={CHAMMI: A benchmark for channel-adaptive models in microscopy imaging},
   booktitle={Advances in Neural Information Processing Systems (NeurIPS) Track on Datasets and Benchmarks},
   year={2023}}
3. Complex Networks

   Deep Distance Sensitivity Oracles

   Davin Jeong, Allison Gunby-Mann, Sarel Cohen, Maximilian Katzmann, Chau Pham, Arnav Bhakta, Tobias Friedrich, and Peter Chin

   In Proceedings of the twelfth International Conference on Complex Networks and Their Applications 2023

   Abstract PDF Bibtex

   One of the most fundamental graph problems is finding a shortest path from a source to a target node. While in its basic forms the problem has been studied extensively and efficient algorithms are known, it becomes significantly harder as soon as parts of the graph are susceptible to failure. Although one can recompute a shortest replacement path after every outage, this is rather inefficient both in time and/or storage. One way to overcome this problem is to shift computational burden from the queries into a pre-processing step, where a data structure is computed that allows for fast querying of replacement paths, typically referred to as a Distance Sensitivity Oracle (DSO). While DSOs have been extensively studied in the theoretical computer science community, to the best of our knowledge this is the first work to construct DSOs using deep learning techniques. We show how to use deep learning to utilize a combinatorial structure of replacement paths. More specifically, we utilize the combinatorial structure of replacement paths as a concatenation of shortest paths and use deep learning to find the pivot nodes for stitching shortest paths into replacement paths.

   @InProceedings{jeongDeepOracles2023,
   author={Davin Jeong and Allison Gunby-Mann and Sarel Cohen and Maximilian Katzmann and Chau Pham and Arnav Bhakta and Tobias Friedrich and Peter Chin},
   title={Deep Distance Sensitivity Oracles},
   booktitle={Proceedings of the twelfth International Conference on Complex Networks and Their Applications},
   year={2023}}

2022

1. A Multi-scale Graph Signature for Persistence Diagrams based on Return Probabilities of Random Walks

   Chau Pham*, Trung Dang*, and Peter Chin

   Preprint (2022)

   Abstract PDF Bibtex

   Persistence diagrams (PDs), often characterized as sets of death and birth of homology class, have been known for providing a topological representation of a graph structure, which is often useful in machine learning tasks. Prior works rely on a single graph signature to construct PDs. In this paper, we explore the use of a family of multi-scale graph signatures to enhance the robustness of topological features. We propose a deep learning architecture to handle this set input. Experiments on benchmark graph classification datasets demonstrate that our proposed architecture outperforms other persistent homology-based methods and achieves competitive performance compared to state-of-the-art methods using graph neural networks. In addition, our approach can be easily applied to large size of input graphs as it does not suffer from limited scalability which can be an issue for graph kernel methods.

   @article{pham2022multiscalegraph,
   title={A Multi-scale Graph Signature for Persistence Diagrams based on Return Probabilities of Random Walks},
   author={Chau Pham and Trung Dang and Peter Chin},
   journal={arXiv preprint arXiv:2209.14264},
   year={2022}}

2020

1. IEEE BigData Cup

   Road Damage Detection and Classification with Detectron2 and Faster R-CNN

   Vung Pham, Chau Pham, and Tommy Dang

   In IEEE International Conference on Big Data (Big Data) 2020

   Abstract PDF Bibtex

   The road is vital for many aspects of life, and road maintenance is crucial for human safety. One of the critical tasks to allow timely repair of road damages is to quickly and efficiently detect and classify them. This work details the strategies and experiments evaluated for these tasks. Specifically, we evaluate Detectron2’s implementation of Faster R-CNN using different base models and configurations. We also experiment with these approaches using the Global Road Damage Detection Challenge 2020, A Track in the IEEE Big Data 2020 Big Data Cup Challenge dataset. The results show that the X101-FPN base model for Faster R-CNN with Detectron2’s default configurations is efficient and general enough to be transferable to different countries in this challenge. This approach results in F1 scores of 51.0% and 51.4% for the test1 and test2 sets of the challenge, respectively. Though the visualizations show good prediction results, the F1 scores are low. Therefore, we also evaluate the prediction results against the existing annotations and discover some discrepancies. Thus, we also suggest strategies to improve the labeling process for this dataset.

   @InProceedings{phamRoadDamage2020,
   title={Road Damage Detection and Classification with Detectron2 and Faster R-CNN},
   author={Vung Pham and Chau Pham and Tommy Dang},
   booktitle={2020 IEEE International Conference on Big Data (Big Data)},
   year={2020},
   pages={5592-5601},
   keywords={Roads; Object detection; Predictive models; Big Data;Data models; Labeling; Task analysis; Object detection; Detectron2; Road damage detection; Faster R-CNN and classification; Transferable learning},
   doi={10.1109/BigData50022.2020.9378027} }
2. IEEE BigData - GTA3 4.0

   Graph Adversarial Attacks and Defense: An Empirical Study on Citation Graph

   Chau Pham, Vung Pham, and Tommy Dang

   In IEEE International Conference on Big Data (GTA3 4.0 Workshop) 2020

   Abstract PDF Bibtex

   This paper details the methodologies and decisions making processes used while developing the attacking and defending models for the Graph Adversarial Attacks and Defense applied to a large citation graph. To handle the large graphs, our attack strategy is twofold: 1) randomly attack the structure first, 2) keep the structure unchanged, then continue the attack on the features using the gradient-based method. On the other hand, the defender is based on 1) filtering and normalizing the feature data, 2) applying the Graph Convolutional Network model, and 3) selecting the models with the highest accuracy and robustness based on our own attacking data. We applied these strategies in KDD Cup 2020 on Graph Adversarial Attacks and Defense dataset. The attacker can drop the accuracy of a surrogate 2-layer Graph Convolutional Network model from 60% to 30% on the test set. Our defending model has 68% accuracy on the validated data and has 89% of the target labels remained the same while adding fake nodes, generated by our attacking method, to the graph.

   @InProceedings{phamGraphAdversarial2020,
   title={Graph Adversarial Attacks and Defense: An Empirical Study on Citation Graph},
   author={Chau Pham and Vung Pham and Tommy Dang},
   booktitle={2020 IEEE International Conference on Big Data (Big Data)},
   year={2020},
   pages={2553-2562},
   keywords={Filtering; Conferences; Neural networks; Decision making; Big Data; Data models; Robustness; graph neural network; graph convolutional network; graph adversarial attacks; graph defense},
   doi={10.1109/BigData50022.2020.9377988} }

2019

1. IEEE BigData Cup

   Solar Flare Prediction Using Two-tier Ensemble with Deep Learning and Gradient Boosting Machine

   Chau Pham, Vung Pham, and Tommy Dang

   In IEEE International Conference on Big Data (Big Data) 2019

   Abstract PDF Bibtex

   This paper describes a machine learning approach to the solar flare prediction competition, a track in IEEE Big Data 2019 Big Data Cup. The competition task is to predict whether or not there is a solar flare event basing on a given time series of solar magnetic field parameters. Our method involves exploring and constructing data-driven machine learning models for the classification task of two imbalanced class labels from time series. Specifically, the investigated models include boosting, logistic regression, multilayer perceptron neural network, and long short-term memory neural network. These models have been successfully deployed and combined in an ensemble framework with two tiers in our final proposed solution for this competition. Our proposed approach ranked at the second place in the competition (the first on the private board and the eleventh on the public board).

   @InProceedings{phamSolarFlare2019,
   title={Solar Flare Prediction Using Two-tier Ensemble with Deep Learning and Gradient Boosting Machine},
   author={Chau Pham and Vung Pham and Tommy Dang},
   booktitle={2019 IEEE International Conference on Big Data (Big Data)},
   year={2019},
   pages={5844-5853},
   keywords={Time series analysis; Task analysis; Neural networks; Machine learning; Biological system modeling; Heating systems; Data visualization; Gradient boosting decision tree; Logistic regression; Multilayer perceptron neural network; Long Short-term memory neural network; Ensemble framework; Solar flare prediction; Multivariate time series},
   doi={10.1109/BigData47090.2019.9006542} }

2018

1. SoICT

   Discord Discovery in Streaming Time Series Based on an Improved HOT SAX Algorithm

   Chau Pham, Bui Minh Duc, and Duong Tuan Anh

   In Proceedings of the Ninth International Symposium on Information and Communication Technology 2018

   Abstract PDF Bibtex

   In this paper, we propose an improved variant of HOT SAX algorithm, called HS-Squeezer, for efficient discord detection in static time series. HS-Squeezer employs clustering rather than augmented trie to arrange two ordering heuristics in HOT SAX. Furthermore, we introduce HS-Squeezer-Stream, the application of HS-Squeezer in the framework for detecting local discords in streaming time series. The experimental results reveal that HS-Squeezer can detect the same quality discords as those detected by HOT SAX but with much shorter run time. Furthermore, HS-Squeezer-Stream demonstrates a fast response in handling time series streams with quality local discords detected.

   @inproceedings{phamDiscordDiscovery2018,
   author = {Chau, Pham Minh and Duc, Bui Minh and Anh, Duong Tuan},
   title = {Discord Discovery in Streaming Time Series based on an Improved HOT SAX Algorithm},
   year = {2018},
   isbn = {9781450365390},
   publisher = {Association for Computing Machinery},
   url = {https://doi.org/10.1145/3287921.3287929},
   doi = {10.1145/3287921.3287929},
   booktitle = {Proceedings of the 9th International Symposium on Information and Communication Technology},
   pages = {24–30},
   keywords = {Streaming time series, clustering, discord discovery},
   }
2. ACOMP

   Discord Detection in Streaming Time Series with the Support of R-Tree

   Chau Pham, Duc Bui Minh, and Anh Duong Tuan

   In International Conference on Advanced Computing and Applications (ACOMP) 2018

   Abstract PDF Bibtex

   Time series discord detection is an important problem in various applications. In this work, we propose a version of Heuristic Discord Discovery (HDD) algorithm, called HDD-MBR, for efficient discord detection in static time series. HDD-MBR employs R-tree as a supporting data structure to arrange two ordering heuristics in HDD. Furthermore, we introduce HDD-MBR-Stream, the application of HDD-MBR in the framework given by Liu et al. for detecting local discords in streaming time series. The experimental results reveal that HDD-MBR can detect the same quality discords as those detected by HOT SAX but with much shorter run time. Moreover, HDD-MBR-Stream demonstrates a fast response in handling time series streams with quality local discords detected.

   @inproceedings{phamDiscordDetection2018,
   author={Pham Minh, Chau and Bui Minh, Duc and Duong Tuan, Anh},
   booktitle={2018 International Conference on Advanced Computing and Applications (ACOMP)},
   title={Discord Detection in Streaming Time Series with the Support of R-Tree},
   year={2018},
   pages={96-103},
   keywords={Time series analysis; Heuristic algorithms; Anomaly detection; Aggregates; Classification algorithms; Monitoring; Detection algorithms; streaming time series; discord discovery; R-Tree},
   doi={10.1109/ACOMP.2018.00023} }