Thuan Pham

Very interesting report from AMSI/AIP regarding declining enrolments and female participation in physics. See our paper below (open access) for a discussion around how the curriculum might factor into this. Georgiou, H., & Abraham, J. (2024). Australian teachers’ perspectives of two distinct physics curriculum paradigms. Physics Education, 59(2), 025017. https://lnkd.in/gjcHkm8j

20

3 Comments

Looking forward to speaking at the Australian Centre for Robotics this week. I will present an efficient approach to time-series modelling with linear recurrent neural networks (LRNNs) - and great to be visiting ACFR again.

40

Our paper entitled "Near OOD Detection for Vision-Language Prompt Learning with Contrastive Logit Score" led by Monash PhD graduate David Jung (now at Oracle) has been published in International Journal of Computer Vision. We proposed a simple score function that can be used to improve near OOD detection performance for trained CLIP-based prompt learning methods, almost for free. Tested with various prompt learning methods. Paper at https://lnkd.in/ev9t4sW4 (open access) Code at https://lnkd.in/ekYRDh7T

17

Excited to announce that our paper on adaptive synchronization of chaotic energy resource systems is now published in the Australian Journal of Electrical and Electronics Engineering. We've introduced a new adaptive sliding mode controller using a backstepping strategy to synchronize two identical chaotic energy resource systems despite external disturbances and different initial conditions. Special thanks to my co-authors Pikaso Pal, Shubhashish Bhakta Vivekananda Mukherjee, Gang Gyoo Jin and Mesfin Megra Rorisab for their contributions. EEE SR University SR University #ChaosTheory #EnergySystems #ControlSystems #Engineering #Research #ElectricalEngineering #Automation #Synchronization

71

In seconds, AI builds proteins to battle cancer and antibiotic resistance; Australian scientists {Monash Uni.) have used Artificial Intelligence (AI) to generate a ready-to-use biological protein that can kill antibiotic resistant bacteria like E. coli. https://lnkd.in/gTMB6c5w

1

🚨🚨New Paper Alert - kaiwen zhang and Amy Gelmi led this work showing how #AFM can uncover how stem cells sense and respond to their mechanical environment during lineage commitment. With Chayla Reeves Kate Fox Mehrnoosh M.. RMIT University RMIT School of Science RMIT Department of Applied Chemistry and Environmental Science 🧪 🔭 https://lnkd.in/gisE38bi

69

Our latest publication focused on applying the active inference framework in driving is now available online (https://lnkd.in/ghEGYHd3)! This work, led by HFML graduate Ran Wei in collaboration with Alfredo Garcia of TAMU, Gustav Markkula of Leeds University, and Waymo (Johan Engström & Matthew O'Kelly), extends our framework to car following. Our findings show that the active inference framework captures a driver behavior more effectively than the gold standard driver process model (IDM) and that it is more interpretable than deep learning approaches (Behavior Cloning) allowing users to understand when and why the model fails. These findings can be used to simulate more accurate human driver behavior in design applications leading to safer automated vehicle control algorithms and advanced driving systems.

54