Portfolio item number 1
Published:
Short description of portfolio item number 1
Posts by Collection
portfolio
Portfolio item number 2
Published:
Short description of portfolio item number 2
publications
Multi-source Transfer Learning for Signal Detection over a Fading Channel with Co-channel Interference
Published in International Conference on Communications (ICC), 2022
For signal detection tasks in wireless communications, most of the existing algorithms either ignore the co-channel interference or treat it as Gaussian noise, which may result in unsatisfactory accuracy when the interference is non-negligible and complex-distributed. When neural networks are motivated in the design of the detectors, difficulty arises in the training due to the fact that there are few accessible pilots in each packet. In this paper, we consider a data-driven detector based on multi-source transfer learning (MSTL) for signal detection in a fading channel with interference. The MSTL detector transfers channel knowledge of previous packets into the latest detection. In particular, we consider a linear combination of the pilots and historical symbols in the distribution space, and design the optimal combination coefficients based on the number of those symbols as well as distributions similarity. Numerical simulations on a Gauss-Markov flat Rayleigh fading channel with co-channel interference validate the advantages of our algorithms, compared with several existing training schemes including directly applying the fully connected deep neural network (FCDNN) and conventional linear minimum mean square error (LMMSE) detector.
Download here
Mitigating Model Poisoning Attacks on Distributed Learning with Heterogeneous Data
Published in International Conference on Machine Learning and Applications (ICMLA), 2023
Gradient-based distributed learning techniques have been essential for machine model training on distributed samples without collecting raw data. However, such learning systems are vulnerable to both internal failures and external attacks. In this work, we study the Byzantine robustness of distributed learning over heterogeneous data for classification tasks, where each worker only contains data samples belonging to some specific classes (aka., label skew) and a fraction of workers are corrupted by a Byzantine adversary to conduct attacks by sending malicious gradients. Existing defenses usually fail under such heterogeneous cases. To remedy this, we propose a gradient decomposition scheme called DeSGD to achieve more robust distributed model training. The key idea for mitigating the impact of data heterogeneity on the Byzantine robustness is to divide the full global gradient into individual gradients of each data class and conduct resilient aggregation in a class-wise manner. The proposed framework can easily integrate existing advanced defense methods and local momentum mechanism. Evaluation results on the Fashion-MNIST dataset with various strong attacks demonstrate the improved robustness of learning over distributed data in the presence of both label skew and attacks.
Download here
White-box Error Correction Code Transformer
Published in Conference on Parsimony and Learning (CPAL), 2025
Error correcting codes (ECCs) play a crucial role in modern communication systems by ensuring reliable data transmission over noisy channels. While traditional algorithms based on belief propagation suffer from limited decoding performance, transformer-based approaches have emerged as powerful solutions for ECC decoding. However, the internal mechanisms of transformer-based approaches remain largely unexplained, making it challenging to understand and improve their performance. In this paper, we propose a White-box Error Correction Code Transformer (WECCT) that provides theoretical insights into transformer-based decoding. By formulating the decoding problem from a sparse rate reduction perspective and introducing a novel Multi-head Tanner-subspaces Self Attention mechanism, our approach provides a parameter-efficient and theoretically principled framework for understanding transformer-based decoding. Extensive experiments across various code families demonstrate that this interpretable design achieves competitive performance compared to state-of-the-art decoders.
Download here
talks
Free Speech Movement
Published:
The Free Speech Movement (FSM) was a massive, long-lasting student protest which took place during the 1964–65 academic year on the campus of the University of California, Berkeley. The Movement was informally under the central leadership of Berkeley graduate student Mario Savio. In fact, the Free Speech Movement is a result of the interaction of history and current affairs, international and American self development. From the east to the west, the wave of world revolution swept the country and shacked the whole world. This talk is divided into 4 parts: background, process, Mario Savio and influence.
This is a presentation in UC Berkeley with Jialin Yang, Qiuyi Zhan and Puzheng Zhang. It is worth mentioning that one of our presenters Qiuyi Zhan is also an intern of the well-known《令人心动的offer第二季》 in 2020-2021.
Mathematics Applied In Our Life: Game Theory and Probability Theory
Published:
This talk introduces some interesting applications of the Game Theory and Probability Theory in our life. To most of us, we don’t want to miss the marriage if we find the person who we think is the best partner.
You may wonder which age to marry achieves the most rationality. Now we can try to solve out this problem. It is supposed that you started dating when you’re 18, and ideally you’d like to be married by the time that you’re 35.
Then you have to choose a percentage that you should reject all your partners to step into your marriage before this percentage of persons you have met. And then you should pick the next person that comes along that is better than everybody you’ve seen before. According to the Optimal Stopping Theory, the best strategy is to reject everybody as serious marriage potential in the first 37％ period. That is about 24.3 years old. So on the mathematical theoretic level, you have the highest probability of meeting the best partner if we start to consider marriage after the 37% time point. Of course, there are many other variables during the period we haven’t consider. If you are interested, you could make a further study of this model. But anyway, it provides us with a rational way to consider our marriage age.
Information in the Representations and Information in the Weights of Deep Learning
Published:
This is a review talk. Effectiveness of deep learning is often ascribed to the ability of deep networks to learn representations. Using established principles from Statistics and Information Theory, we introduce the desirable properties of representations and the relationships between them. It is shown that invariance to nuisance factors in a deep neural network is equivalent to information minimality of the learned representation. On the other hand, we want to find parameters (weights) that yield good generalization, so we focus on the information in the weights which can be controlled by implicit or explicit regularization. With some additional assumptions, we get a connection between information in the representations and information in the weights.
Multi-source Transfer Learning for Signal Detection over a Fading Channel with Co-channel Interference
Published:
For signal detection tasks in wireless communications, most of the existing algorithms either ignore the co-channel interference or treat it as Gaussian noise, which may result in unsatisfactory accuracy when the interference is non-negligible and complex-distributed. When neural networks are motivated in the design of the detectors, difficulty arises in the training due to the fact that there are few accessible pilots in each packet. In this paper, we consider a data-driven detector based on multi-source transfer learning (MSTL) for signal detection in a fading channel with interference. The MSTL detector transfers channel knowledge of previous packets into the latest detection. In particular, we consider a linear combination of the pilots and historical symbols in the distribution space, and design the optimal combination coefficients based on the number of those symbols as well as distributions similarity. Numerical simulations on a Gauss-Markov flat Rayleigh fading channel with co-channel interference validate the advantages of our algorithms, compared with several existing training schemes including directly applying the fully connected deep neural network (FCDNN) and conventional linear minimum mean square error (LMMSE) detector.
A Mathematical Framework of Multi-source Transfer Learning
Published:
This talk introduces a theoretical framework for multi-source transferability covering distance, sample sizes, and task complexity at the same time. It extends to continuous data via training neural networks. Futhermore, a consistent algorithm is proposed that works in applications.
teaching
Teaching experience 1
Undergraduate course, University 1, Department, 2014
This is a description of a teaching experience. You can use markdown like any other post.
Teaching experience 2
Workshop, University 1, Department, 2015
This is a description of a teaching experience. You can use markdown like any other post.