ZhiruoZhou

Professor Kuo Delivered Keynote at PCS 2022 on Green Coding

The Picture Coding Symposium (PCS) is an international forum devoted to advances in visual data coding. Established in 1969, it has the longest history of any conference in this area. The 36th event in the series, PCS 2022, was held from December 7-9 in San Jose, California, USA, the heart of Silicon Valley and the cultural and technological epicenter of Northern California.  The conference venue was the San Jose Hilton hotel.

MCL Director, Professor C.-C. Jay Kuo, was invited to deliver a keynote speech on green coding on Dec. 7. The abstract of his keynote was “Green Coding: Low-Complexity Learning-based Image/Video Coding.” The abstract of his talk was:

“Deep-learning-based coding (or deep coding in short) has attracted much attention in recent years due to its superior rate-distortion (RD) performance. Yet, its huge computational complexity and model sizes are of concern in practical applications.  An alternative learning-based coding, called green coding, has been intensively studied in my lab for the last two and half years. Green coding targets a model size that is significantly smaller than that of deep coding. Furthermore, it has much lower decoding complexity than today’s advanced codecs, such as HEVC and VVC. It is particularly attractive for mobile devices. Green coding uses multi-grids to capture short-, mid-, and long-range correlations in images and adopts vector quantization (VQ) to leverage correlations between images. Extensive experiments are conducted to demonstrate the high RD performance and low complexity of green image coding. Its generalization to green video coding will also be discussed.”

Besides, Professor Kuo visited San Clara University on December 6. Hosted by Professor Nam Ling, he and Professor Chia-Wen Lin of National Tsinghua University gave two lectures, which were events of the US local chapter [...]

By |December 11th, 2022|News|Comments Off on Professor Kuo Delivered Keynote at PCS 2022 on Green Coding|

MCL Research on Generated Samples Quality Assessment

Despite prolific work on evaluating generative models, little research has been done on the quality evaluation of an individual generated sample. To address this problem, a lightweight generated sample quality evaluation (LGSQE) method is proposed in this work. In the training stage of LGSQE, a binary classifier is trained on real and synthetic samples, where real and synthetic data are labeled by 0 and 1, respectively. In the inference stage, the classifier assign soft labels (ranging from 0 to 1) to each generated sample. The value of soft label indicates the quality level; namely,the quality is better if its soft label is closer to 0. LGSQE can serve as a post-processing module for quality control. Furthermore, LGSQE can be used to evaluate the performance of generative models, such as accuracy, AUC, precision and recall, by aggregating sample-level quality. Experiments are conducted on CIFAR-10 and MNIST to demonstrate that LGSQE can preserve the same performance rank order as that predicted by the Frechet Inception Distance (FID) but with significantly lower complexity.

Fig. 1 shows the pipeline of the proposed method. The LGSQE method consists of three cascaded modules:

Module 1: Representation Learning. effective local and global representations of images are learned based upon PixelHop framework.

Module 2: Discriminant Feature Test (DFT). Use DFT to choose powerful features from large numbers of representations obtained from Module 1 against a particular task.

Module 3: Binary Classification for Evaluation. We partition the real/generated data into training and testing sets. A binary classifier is trained on the union of real and generated training samples, which are labeled with “0” and “1”, respectively. The classifier assigns a soft score, to each testing sample as the sample quality index.

Fig. 2 shows the evaluation of generated [...]

By |December 4th, 2022|News|Comments Off on MCL Research on Generated Samples Quality Assessment|

MCL Thanksgiving Luncheon

MCL had the annual Thanksgiving Luncheon at Shiki Seafood Buffet on November 24, 2022. The Thanksgiving Luncheon has been a tradition of MCL for more than 20 years. It’s a good chance for the whole group to gather and have a lunch together as a warm and happy family. All of us enjoyed the delicious food and the wonderful time chatting with each other. It’s also a good opportunity to have a rest after a busy semester and get connected with other people. Thank Professor Kuo for holding this event and thank Xuejing for organizing it.

Happy Thanksgiving to everyone!

By |November 27th, 2022|News|Comments Off on MCL Thanksgiving Luncheon|

MCL Genealogical Ancestry Series: Johann Georg Büsch

Hong-Shuo Chen studied Johann Georg Büsch, the mathematician in MCL genealogy, and shared his study with MCL members in the pre-seminar sharing on November 14, 2022. Johann Georg Büsch was born on January 3, 1728, at Alten-Weding in Hanover and died on August 5, 1800, in Hamburg. He is a German mathematics teacher and writer on statistics and commerce. He is the 14th ancestor of Professor Kuo.

When he was three years old, he came to Hamburg with his parents. He studied at the Georg-August-University of Gottingen and the Martin Luther University of Halle Wittenberg. After he graduates, he becomes a mathematics professor in the Hamburg gymnasium. He wrote a lot of books related to commerce and political economy.  As a math teacher, he mentored and helped the young Johann Elert Bode, who later became a famous astronomer. Johann Elert Bode calculated the orbit of Uranus and recommended the planet’s name.

Johann Georg Büsch married Margarete Augusta Schwalb in 1759, and they had 5 sons and 5 daughters. Two sons became the merchant. He also owned a library of 3200 math books and some scientific instruments. Besides his contribution to math, he established an association for the promotion of art and industry and the foundation of a school of trade. Both of them became one of the most noted establishments of their class in the world under his direction. Before his death, Büsch was almost totally blind. However, he was still working effortlessly until the last day of his life.

 

References: https://en.wikipedia.org/wiki/Johann_Georg_B%C3%BCsch

By |November 20th, 2022|News|Comments Off on MCL Genealogical Ancestry Series: Johann Georg Büsch|
  • Permalink Gallery

    MCL’s Word Embedding Paper Won the Sadaoki Furui Prize Paper Award

MCL’s Word Embedding Paper Won the Sadaoki Furui Prize Paper Award

Congratulations to Bin Wang, Angela Wang, Jessica Chen, Yun-Cheng Wang, and Professor Jay Kuo, for receiving the 2022 Sadaoki Furui Prize Paper Award at Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC 2022) in Chiang Mai, Thailand for the work:

“Bin Wang, Angela Wang, Fenxiao Chen, Yuncheng Wang, and C-C. Jay Kuo. “Evaluating Word Embedding Models: Methods and Experimental Results.” APSIPA Transactions on Signal and Information Processing 8 (2019).”

The APSIPA Sadaoki Furui Prize Paper Award is awarded at APSIPA ASC each year based on a selection from the papers published in the preceding five years on APSIPA Transactions on Signal and Information Processing. It is a great honor to receive this award and support from the APSIPA society.

The paper focuses on word embedding methods and their evaluation. Word embedding, also known as word representation, is a powerful tool that is widely used in modern natural language processing (NLP) and cross-subject areas like knowledge representation and multi-modality learning. The goal of word embedding is to learn vector representations for words commonly served as the first step to most NLP applications.

There are several major contributions to the work. First, an in-depth discussion on what properties serve good word embedding and word embedding evaluators is presented. Then, the paper contains not only comprehensive surveys on word embedding and evaluation methods but also extensive experimental evaluations of these models. The evaluation methods are categorized as intrinsic and extrinsic ones. A comprehensive correlation study between them is analyzed for the first time. The paper inspires a series of research works and the citation number has reached around 200 times in three years.

Our lab continuously contributes to the field of natural language processing (NLP) including representation learning, [...]

By |November 13th, 2022|News|Comments Off on MCL’s Word Embedding Paper Won the Sadaoki Furui Prize Paper Award|
  • Permalink Gallery

    Professor Kuo Delivered an Invited Lecture at Caltech’s Keller Colloquium

Professor Kuo Delivered an Invited Lecture at Caltech’s Keller Colloquium

The Keller Colloquium is a distinguished lecture series for the CMS (Computing and Mathematical Sciences) department at the California Institute of Technology (Caltech). The CMS department includes Applied Math, Control, and Computer Science programs. A committee of students and faculty chooses the speakers across these areas. Professor Kuo was invited to lecture during the Fall Quarter of the 2022-23 academic year.

Professor Kuo visited Caltech on October 31 (Monday). He met Professor P. P. Vaidyanathan and Professor Thomas Yizhao Hou before his seminar at 4 pm. The title of his lecture was “Green Learning: Methodology, Examples, and Outlook,” with the following abstract:

“The rapid advances in artificial intelligence in the last decade are primarily attributed to the wide applications of deep learning (DL). Yet, the high carbon footprint yielded by larger DL networks is a concern to sustainability. Green learning (GL) has been proposed as an alternative to address this concern. GL is characterized by low carbon footprints, small model sizes, low computational complexity, and mathematical transparency.  It offers energy-effective solutions in cloud centers and mobile/edge devices.  It has three main modules: 1) unsupervised representation learning, 2) supervised feature learning, and 3) decision learning. GL has been successfully applied to a few applications. This talk provides an overview of the GL solution, its demonstrated examples, and its technical outlook. The connection between GL and DL will also be discussed.”

The Lecture was well attended. People showed interest in this new machine-learning paradigm.

By |November 6th, 2022|News|Comments Off on Professor Kuo Delivered an Invited Lecture at Caltech’s Keller Colloquium|
  • Permalink Gallery

    Professor Kuo Delivered Green Learning Tutorial at ICIP-2022

Professor Kuo Delivered Green Learning Tutorial at ICIP-2022

The 29th IEEE International Conference on Image Processing (IEEE ICIP) was held in Bordeaux, France on October 16-19, 2022. The IEEE ICIP is the world’s largest and most comprehensive technical conference focused on image and video processing and computer vision. Professor C.-C. Jay Kuo, the Director of USC Media Communications Lab (MCL), gave a tutorial on “Green Learning: Methodologies and Applications” in the afternoon of October 16 (Sunday), 2-5:30 pm. Here is the description of the tutorial.

“There has been a rapid development of artificial intelligence and machine learning technologies in the last decade. The core lies in many annotated training data and deep learning networks. Representative deep learning networks include the convolutional neural network, the recurrent neural network, the long short-term memory network, the transformer, etc. Although deep learning networks have significantly impacted applications such as computer vision, natural language processing, autonomous driving, robotics navigation, etc., they have several inherent shortcomings. They are mathematically intractable, vulnerable to adversarial attacks, and demand a massive amount of annotated training data. Furthermore, their training is computationally intensive because of the use of backpropagation for end-to-end network optimization.”

“There is an emerging concern that deep learning technologies are not friendly to the environment since their carbon footprint threatens global warming and climate change. As sustainability has become critical to human civilization, one priority in science and engineering is to preserve our environment for future generations. In artificial intelligence, it is urgent to investigate new learning paradigms that are competitive with deep learning in performance yet with a significantly lower carbon footprint. Professor C.-C. Jay Kuo has worked towards this goal since 2014. He has published a sequence of influential papers along this direction (see the recent publication list) and [...]

By |October 30th, 2022|News|Comments Off on Professor Kuo Delivered Green Learning Tutorial at ICIP-2022|

MCL Research on Green Knowledge Graph Completion

Knowledge graphs (KGs) store human knowledge in a graph-structured format, where nodes and edges denote entities and relations, respectively. Most KGs, such as wikidata [1], suffer from the incompleteness problem; namely, a large number of factual triples are missing, leading to performance degradation in downstream applications. Thus, there is growing interest in developing KG completion (KGC) methods to solve the incompleteness problem by inferring undiscovered factual triples based on existing ones.

Prior KGC work focuses on learning embeddings for entities and relations through a simple score function. Yet, a higher-dimensional embedding space is usually required for a better reasoning capability, which leads to larger model size and hinders applicability to real-world problems (e.g., large-scale KGs or mobile/edge computing). A lightweight modularized KGC solution, called GreenKGC, is proposed to address this issue. GreenKGC consists of three modules: representation learning, feature pruning, and decision learning, to extract discriminant KG features and make accurate predictions on missing relationships using classifiers and negative sampling. The overview of the model is shown in Fig. 1.

Experimental results demonstrate several advantages of GreenKGC. First, it only requires a low-dimensional space (e.g. d = 8) to achieve achieve competitive or even better performance against high-dimensional models with much smaller model sizes. Second, as compared with other classification-based methods, it requires a shorter inference time and provides better performance. Third, the feature pruning module is 20x faster than knowledge distillation methods in training powerful low-dimensional features. A comparison of GreenKGC and other KGC models under different dimensions is given in Fig. 2.

-By Yun Cheng Wang

 

Reference:

[1] Vrandečić, Denny, and Markus Krötzsch. “Wikidata: a free collaborative knowledgebase.” Communications of the ACM 57.10 (2014): 78-85.

By |October 23rd, 2022|News|Comments Off on MCL Research on Green Knowledge Graph Completion|

MCL Genealogical Ancestry Series: Carl Fredrich Gauss

In the MCL genealogy, Carl Friedrich Gauss is a shining star. Min Zhang studied his unusual life and shared with MCL members in the pre-seminar sharing on October 03, 2022. Gauss is the greatest mathematician since antiquity, known for his contributions to number theory, proving the fundamental theorem of algebra, deriving the function representation of normal distribution, his contribution to the theory of magnetism and being PhD advisor to Richard Dedekind and Bernhard Riemann.

Carl Friedrich Gauss was born on April 30, 1777, in Brunswick in the Duchy of Brunswick-Wolfenbüttel which now part of Lower Saxony, Germany. His parents were poor, working-class citizens. His mother was illiterate and never recorded his birthdate, remembering only that was a Wednesday, eight days before the Feast of the Ascension. Gauss figured out his birthday by deriving the date of Easter by himself. There are some interesting anecdotes about Gauss when he was a child. Gauss is child prodigy, he was said to have corrected an error in this father’s payroll calculations at the age of 3, he dazzled his schoolteachers by quickly summing up the integers from 1 to 100 to be 5050 at the age of 7 and he was already criticizing Euclid’s geometry at the age of 12.

Gauss’s intellectual abilities attracted the attention of the Duke of Brunswick, the Duke supported his study and life since he was 14 years old until the Duke passed away in 1806. With the funding from the Duke, Gauss studied at the Collegium Carolinum (now Braunschweig University of Technology) from 1792 to 1795. Then, he got a bachelor’s degree at the University of Göttingen in 1798 when he was 22. Gauss was very productive in 1796. He advanced modular arithmetic [...]

By |October 16th, 2022|News|Comments Off on MCL Genealogical Ancestry Series: Carl Fredrich Gauss|

MCL Genealogical Ancestry Series: Nicolaus Copernicus

Xuejing Lei studied on Nicolaus Copernicus, the first mathematician in MCL genealogy, and shared her study with MCL members in the pre-seminar sharing on September 12, 2022. Nicolaus Copernicus was born in 1473 in Toruń, Royal Prussia, Poland and died in 1543 aged 70. He was a Renaissance polymath, who made contributions in a wide variety of fields including astronomy, canon law, economics, mathematics, etc. He is best known for Heliocentrism, Quantity theory of money and Gresham–Copernicus law.

Nicolaus Copernicus was born in a powerful family. His father was a well-to-do merchant who dealt in copper, and died about 1483. His mother was the daughter of a wealthy Toruń patrician and city councilor, deceased after 1495. After his father’s death, his maternal uncle, Lucas Watzenrode the Younger, took the little boy under his wing and saw to his education and career. Lucas formed close relations with three successive Polish monarchs Watzenrode and many rulers. He came to be considered the most powerful man in Warmia, and his wealth, connections and influence allowed him to secure Copernicus’s education and career as a canon at Frombork Cathedral.

Nicolaus Copernicus’s study in University of Kraków (now Jagiellonian University) gave him a thorough grounding in the mathematical astronomy and initiated his analysis of logical contradictions in the two “official” systems of astronomy — Aristotle’s theory of homocentric spheres, and Ptolemy’s mechanism of eccentrics and epicycles. He then went to Italy and studied in University of Bologna for 4 years and University of Padua for 2 years, and obtained his Doctoral degree of Canon Law in University of Ferrara in 1503.

Although Nicolaus Copernicus was best known to his contemporaries as a doctor and the Canon of Frauenburg Cathedral, he is best [...]

By |October 9th, 2022|News|Comments Off on MCL Genealogical Ancestry Series: Nicolaus Copernicus|