Zhe Hou

Hi everyone! I’d like to share one of my latest publications on performing Bayesian inference using ABC (Approximate Bayesian Computation), coauthored with David Frazier from Monash University. In this work, we introduce a new method for diagnosing model misspecification in cases where inference is done using ABC. Unlike previous methods, ours is based on the asymptotic properties of ABC. We show, both theoretically and through examples, that our method can reliably detect model misspecification and performs well with finite sample sizes, often outperforming existing approaches. Linnk: https://lnkd.in/e8d-YvKv

63

1 Comment

*Open PhD positions - high-dimensional quantum computing in silicon* My group at #unsw has several open PhD positions, to conduct research on the encoding of quantum information in high-dimensional nuclear spins in silicon. These atomic-size systems can be used to efficiently encode logical qubits. Spread the word! https://lnkd.in/gNpQQPXq #PhD #quantumcomputing #quantum

137

3 Comments

Researchers Slash Quantum Circuit Complexity for Machine Learning Researchers from CSIRO and the University of Melbourne have made significant progress in quantum machine learning, a field aimed at achieving quantum advantage to outperform classical machine learning. Their work demonstrates that quantum circuits for data encoding in quantum machine learning can be greatly simplified without compromising accuracy or robustness. The team's results, validated through both simulations and experiments on IBM quantum devices, show that their innovative encoding methods reduced circuit depth by a factor of 100 on average compared to traditional approaches while achieving similar classification accuracies. The study, published in Intelligent Computing, introduces three encoding methods - matrix product state, genetic, and variational algorithms - that approximate the quantum state of the data using much shallower circuits. These methods maintained classification accuracy on the MNIST image dataset and two others while enhancing resilience against adversarial data manipulation. The work aligns with broader goals in quantum machine learning to build efficient, reliable quantum models for areas such as image recognition, cybersecurity, and complex data analysis. https://lnkd.in/e8ndUyRS

21

👏 Congratulations to Professor Xiaohui Liu and Professor Kezhi Wang from the Department of Computer Science who have been named on the Clarivate Highly Cited Researchers List 2024. This is Prof Liu's (pictured left) 11th year named on the 'who's who' list of the top 1% of published academics. He has authored well over 300 papers in research topics including artificial intelligence, data science and optimisation. Prof Wang's research on wireless communication and machine learning puts him amongst the most influential in the field for a second year. The highly cited researchers’ names are drawn from the publications that rank in the top 1% by citations for field and for publication year in the Web of Science citation index. We're so proud to see the work of our academics recognised. 🙌 #Clarivate #HighlyCited #ResearchExcellence #ComputerScience

118

10 Comments

“Quantum computing, communications and sensing could add $6.1 billion to Australia’s GDP”, “the Australian quantum sector could create around 8,700 jobs”, “Queensland: The Quantum State”, so they said. I did a quick search on LinkedIn. I found the following companies in Australia with relevant positions: - Quantum Brilliance (1 position) - Archer (1 position) - Silicon Quantum Computations (3 positions, mostly engineering rather than quantum physics) - Q-CTRL (1 position, posted for a few months) Plus about 3 open postdoctoral positions in various universities and CSIRO. where are all these thousands jobs for quantum physicists?? My recent outstanding phd students (all local north QLD  undergraduate ->honour ->PhD) have moved to Europe for high-quality research positions in quantum physics.  This made me question whether I am giving accurate advice to young, talented Queensland minds when I encourage them to study quantum physics by promising bright career opportunities home.

1

1 Comment

Here's the pulse of Australia's quantum technologies projects! A very well written and illustrated status report of Australian quantum technology projects. I am inspired by the success of the broad and diverse quantum communities. This report helps me better understand where the cool projects of tomorrow are happening. it's a great time to be in project management! #QuantumAwareness #PQC #CyberSecurity #ProjectManagement #PMO #ProjectManager Bond University #SustainableCareer

5

I am happy our paper, "Deep Reinforcement Learning Approach to Portfolio Optimization in the Australian Stock Market" has been published successfully in the AI, Computer Science and Robotics Technology (ACRT) Journal. Congratulations & well done Weiye Wu👏 👏 👏 ! Below is a copy of our abstract: The future of portfolio management is evolving from relying on human expertise to incorporating artificial intelligence techniques. Traditional techniques such as fundamental and technical analysis will eventually be replaced by more sophisticated deep reinforcement learning (DRL) algorithms. However, it is still a long way from designing a profitable strategy in the complex and dynamic stock market. While previous studies have focused on the American stock market, this paper applies two DRL algorithms, the proximal policy optimization (PPO) and the advantage actor–critic (A2C), to trade the constituent stocks of the Australian Securities Exchange 50 (ASX50) Index. This paper also incorporates a weighted moving average into the action space and introduces a transaction threshold to help agents minimize trivial trades that lead to high transaction costs. The results are presented and benchmarked against the ASX50 Index. The A2C agent was better at following trends and had the higher upside potential but can suffer from more severe damage during bearish markets. On the other hand, the PPO agent had the lowest annual volatility and the highest maximum drawdown, which is more helpful in a bearish or volatile market. See: https://lnkd.in/gXyaCFVA @DataAnalytic #deepreinforcement #learning #portfolio #optimization #optimization, #analytics #algorithms #prediction #stocks #trading #AI #ASX #stockmarket

28

2 Comments

This week as my feed filled with posts about OpenAI's latest GPT, they all seemed to skip a fundamental aspect of AI: the hardware - aka the high performance computing resources that are required to train and these models. The sustainability of AI is a topic that doesn't get as much discourse as AI itself - which is odd since without the infrastructure, AI models simply couldn't exist. So I was glad to contribute to a panel held by the The Australian Academy of Science where we discussed how Australia could improve our high-performance computing capabilities. The report is now released - you can access it via the link below but here's a TLDR: - Need for a National Strategy: Currently, there is no national strategy for the next generation of HPCs in Australia. Given the escalating needs for research and the need to keep pace with global advancements, such a strategy is critical. - Bolstering AI Capabilities: Australia’s research HPC also bolsters our AI capabilities. Without computing power on our own shores, Australia’s AI capability relies on other nations. - Potential of Quantum and Cloud Computing: Despite their potential, quantum computing and cloud computing cannot replace traditional HPC functions for scientists in the near term. - Ideal Host for Future HPC: Australia’s stability, connectivity, and renewable energy capacity position us as an ideal host for future regional HPC, especially an Asia-Pacific Tier-0 exascale facility. - Securing Sovereign Capability: Securing sovereign capability depends on both investment in infrastructure and investment in software readiness programs. Similarly, the role of training, education, data management and AI is crucial to support future growth. Link to full report: https://lnkd.in/gms43cV8 #HPC #AI #quantum #exascale

18

2 Comments

I am pleased to share our article published in APL Quantum. This work is the result of my master's research, in which we investigated a method for reconstructing quantum states of multiple qubits. We used IBM quantum computers to perform experiments (IBM, IBM Quantum) and Qiskit. In the context of quantum computing, state reconstruction is a crucial tool. It allows us, for example, to determine the input or output states of a given algorithm, or even to characterize the logical operations that make up the algorithm. Here is the link to the paper:

26

1 Comment

Paper: Physics-informed neural network reconciles Australian displacements and tectonic stresses By Thomas Poulet & Pouria Behnoudfar Abstract Stress orientation information is invaluable to evaluate active tectonic forces within the Earth’s crust. The global dataset provided by the World Stress Map offers a rich resource of stress indicators, facilitating the calibration of mechanical models to extract complete stress and displacement fields. However, traditional inversion processes are hampered by the manual tuning of geomechanical properties and boundary conditions to reconcile simulations with observations. In this study, we introduce ML-SEISMIC (machine learning for stress estimation integrating satellite image and computational modelling), a physics-informed deep neural network approach to autonomously align stress orientation data with an elastic model. It nearly completely bypasses the need for explicit boundary condition inputs and yields comprehensive distributions of material properties, displacements, and stress tensors. Application of this methodology to Australia, coupled with precise global navigation satellite systems observations, unveils a robust and scale-independent interpolation framework. Additionally, it pinpoints regions where stress orientation reinterpretation is warranted. Our results present a streamlined yet powerful process, offering a substantial leap forward in geodynamic investigations. This approach promises to unify velocity and stress orientation observations with physical models, ushering in a new era of insights into Earth’s dynamic processes #Physicsinformed #neuralnetwork #AI #Australia #tectonic #stresses #geomechanics

12

3 Comments

(cont. from https://lnkd.in/gRfY-jRt) SPIRE paper: Online Computation of String Net Frequency with Peaker Guo (University of Melbourne), Tony Wirth (University of Sydney, University of Melbourne), Justin Zobel (University of Melbourne) The net frequency (NF) of a string in a text is the number of occurrences of the string in the text with unique left and right extensions. NF was originally introduced for Chinese NLP tasks but there is a lack of understanding of the properties of NF and the absence of efficient algorithms for computing NF. At CPM 2024, we showed that NF is combinatorially interesting and how NF can be computed efficiently in the offline setting. For many applications, however, facilitating the computation in an online manner is highly desirable. We are the first to solve the problem in the online setting, and we do so in optimal time.

6

Google Quantum AI demonstrated quantum memory enhanced with access to conjugate quantum states provides exponential advantages in learning tasks such as shadow tomography, providing more efficient information extraction from quantum systems with minimal quantum memory. Researchers at the UNSW demonstrated the first provable quantum entanglement between two phosphorus atoms in silicon, highly relevant for scaling quantum computing using existing semiconductor technology. A study from the Fujitsu Research Quantum Laboratory introduces a resource-efficient equivariant quantum convolutional neural network (sp-QCNN) that leans on general symmetries to improve trainability, generalization, and measurement efficiency for quantum machine learning on near-term quantum devices. Plus, insights into quantum phase transitions, a new photonic quantum architecture, Maxwell’s demon, and more. Read every qubit 👉 https://lnkd.in/gvCF--3g #QuantumComputing #QuantumTechnology #Innovation

15

Led by Associate Professor Giuseppe M. J. Barca, a research team at the University of Melbourne has achieved the first quantum simulation of biological systems at a scale necessary to accurately model drug performance. Utilizing the Frontier supercomputer, this groundbreaking software can predict the chemical reactions and physical properties of molecular systems with unprecedented precision. This breakthrough enables the simulation of drug behavior with an accuracy that rivals physical experiments, essential for designing new therapeutics. The collaboration with Oak Ridge National Laboratory, AMD, and QDX marks a significant advancement in computational chemistry. This technology promises to accelerate drug discovery, making it faster and more cost-effective, and opening doors to treating previously untreatable diseases. QDX, co-founded by Barca, is already leveraging these high-performance quantum simulations to transform therapeutics design globally. https://lnkd.in/g3YZ46g2 #Quantum #Pharmaceuticals #IP #Patents #VC #DeepTech

1

Please consider submitting a paper to a Special Issue on Human-Centric Software Engineering and Cyber Security for Intelligent Systems Link: https://lnkd.in/gEerp9Hn Submission Deadline: 28 February 2025 #humancentric #cybersecurity #softwareengineering #intelligentsystems Xiao Liu Yuan Tian Karen Renaud

35

Our paper entitled "Relation-aware heterogeneous graph network for learning inter-modal semantics in textbook question answering” just accepted in #TNNLS (IEEE Transactions on Neural Networks and Learning Systems) https://lnkd.in/g_XqrhZR

1

Researchers at the University of Sydney have developed a 3D quantum error-correction architecture that improves both the reliability and efficiency of quantum computing. 🚨 This innovative design reduces the number of physical qubits required for error correction, freeing up more qubits for processing and enabling a more compact “quantum hard drive.” By scaling to correct more errors per qubit, this approach marks a significant step forward in error management, with potential applications in cryptography, complex simulations, and beyond. 👇🏻 https://lnkd.in/e8XmyD2X #QuantumComputing #UniversityOfSydney #QuantumInnovation #ErrorCorrection

10