Cicada Innovations has acquired custodianship of Australia's longest-running annual deep technology conference Tech23 as the capital raised by their deep tech companies eclipses $5.7 billion. Tech23 and Cicada Innovations have always been united by a shared mission to celebrate and support Australian deep tech innovators, with both organisations boasting a lengthy track record of nurturing some of Australia's most successful companies: For 13 years, Tech23 has showcased and championed over 300 of Australia's most promising high-growth deep tech companies, while providing a mechanism for venture capital, other investors, media, advisors, policy makers, and potential local and global customers to learn about and connect with these companies. The event, planned for July 2023, will ask big questions about how we build the future – from sustainable cities to game-changing climate solutions and completely new paradigms of human health. Creating a space to connect and spark meaningful conversations about the ideas, technology, and innovators that can propel humanity into a better tomorrow. Cicada will take a proactive approach to finding the next 23 startups, setting out on a national road show in early 2023 to engage with local ecosystems and unearth emerging deep tech voices and give them national platform and vital exposure and support.

The SKA pulsar search pipeline will be used for real time detection of pulsars. Modern radio telescopes such as SKA will be generating petabytes of data in their full scale of operation. Hence experience-based and datadriven algorithms are being investigated for applications such as candidate detection. Here we describe our findings from testing a state of the art object detection algorithm called Mask R-CNN to detect candidate signatures in the SKA pulsar search pipeline. We have trained the Mask R-CNN model to detect candidate images. A custom semi-auto annotation tool was developed and investigated to rapidly mark the regions of interest in large datasets. We have used a simulation dataset to train and build the candidate detection algorithm. A more detailed analysis is planned. The paper presents details of this initial investigation highlighting the future prospects.

Construction of dictionaries using nonnegative matrix factorisation (NMF) has extensive applications in signal processing and machine learning. With the advances in deep learning, training compact and robust dictionaries using deep neural networks, i.e., dictionaries of deep features, has been proposed. In this study, we propose a probabilistic generative model which employs a variational autoencoder (VAE) to perform nonnegative dictionary learning. In contrast to the existing VAE models, we cast the model under a statistical framework with latent variables obeying a Gamma distribution and design a new loss function to guarantee the nonnegative dictionaries. We adopt an acceptance-rejection sampling reparameterization trick to update the latent variables iteratively. We apply the dictionaries learned from VAE-NMF to two signal processing tasks, i.e., enhancement of speech and extraction of muscle synergies. Experimental results demonstrate that VAE-NMF performs better in learning the latent nonnegative dictionaries in comparison with state-of-the-art methods.

Image-text contrastive learning models such as CLIP have demonstrated strong task transfer ability. The high generality and usability of these visual models is achieved via a web-scale data collection process to ensure broad concept coverage, followed by expensive pre-training to feed all the knowledge into model weights. Alternatively, we propose REACT, REtrieval-Augmented CusTomization, a framework to acquire the relevant web knowledge to build customized visual models for target domains. We retrieve the most relevant image-text pairs (~3% of CLIP pre-training data) from the web-scale database as external knowledge, and propose to customize the model by only training new modualized blocks while freezing all the original weights. The effectiveness of REACT is demonstrated via extensive experiments on classification, retrieval, detection and segmentation tasks, including zero, few, and full-shot settings. Particularly, on the zero-shot classification task, compared with CLIP, it achieves up to 5.4% improvement on ImageNet and 3.7% on the ELEVATER benchmark (20 datasets).

In recent years, neural models learned through self-supervised pretraining on large scale multilingual text or speech data have exhibited promising results for underresourced languages, especially when a relatively large amount of data from related language(s) is available. While the technology has a potential for facilitating tasks carried out in language documentation projects, such as speech transcription, pretraining a multilingual model from scratch for every new language would be highly impractical. We investigate the possibility for adapting an existing multilingual wav2vec 2.0 model for a new language, focusing on actual fieldwork data from a critically endangered tongue: Ainu. Specifically, we (i) examine the feasibility of leveraging data from similar languages also in fine-tuning; (ii) verify whether the model's performance can be improved by further pretraining on target language data. Our results show that continued pretraining is the most effective method to adapt a wav2vec 2.0 model for a new language and leads to considerable reduction in error rates. Furthermore, we find that if a model pretrained on a related speech variety or an unrelated language with similar phonological characteristics is available, multilingual fine-tuning using additional data from that language can have positive impact on speech recognition performance when there is very little labeled data in the target language.

Social reasoning necessitates the capacity of theory of mind (ToM), the ability to contextualise and attribute mental states to others without having access to their internal cognitive structure. Recent machine learning approaches to ToM have demonstrated that we can train the observer to read the past and present behaviours of other agents and infer their beliefs (including false beliefs about things that no longer exist), goals, intentions and future actions. The challenges arise when the behavioural space is complex, demanding skilful space navigation for rapidly changing contexts for an extended period. We tackle the challenges by equipping the observer with novel neural memory mechanisms to encode, and hierarchical attention to selectively retrieve information about others. The memories allow rapid, selective querying of distal related past behaviours of others to deliberatively reason about their current mental state, beliefs and future behaviours. This results in ToMMY, a theory of mind model that learns to reason while making little assumptions about the underlying mental processes. We also construct a new suite of experiments to demonstrate that memories facilitate the learning process and achieve better theory of mind performance, especially for high-demand false-belief tasks that require inferring through multiple steps of changes.

Vocal fatigue refers to the feeling of tiredness and weakness of voice due to extended utilization. This paper investigates the effectiveness of neural embeddings for the detection of vocal fatigue. We compare x-vectors, ECAPA-TDNN, and wav2vec 2.0 embeddings on a corpus of academic spoken English. Low-dimensional mappings of the data reveal that neural embeddings capture information about the change in vocal characteristics of a speaker during prolonged voice usage. We show that vocal fatigue can be reliably predicted using all three kinds of neural embeddings after only 50 minutes of continuous speaking when temporal smoothing and normalization are applied to the extracted embeddings. We employ support vector machines for classification and achieve accuracy scores of 81% using x-vectors, 85% using ECAPA-TDNN embeddings, and 82% using wav2vec 2.0 embeddings as input features. We obtain an accuracy score of 76%, when the trained system is applied to a different speaker and recording environment without any adaptation.

Cipia an AI computer vision in-cabin automotive solutions provider announced that the company had secured a design win with an additional Chinese OEM together with Tier 1 HiRain. Cipia's Driver Sense Driver Monitoring System (DMS) will be incorporated into a pickup truck model that will be sold in South America, Australia and New Zealand. The model is expected to start production in 2023. With this announcement, Cipia has now been awarded 29 design wins across 6 car manufacturers. Yehuda Holtzman, CEO of Cipia, said "The selection of Cipia's DMS by a new OEM and Tier 1 demonstrates again the reliability and quality of our technology, and our ability to deliver safer and better driving experiences. We are continuing to push and expand our collaborations with new OEMs, focusing on the US, European and Chinese markets."

Working set size estimation (WSS) is of great significance to improve the efficiency of program executing and memory arrangement in modern operating systems. Previous work proposed several methods to estimate WSS, including self-balloning, Zballoning and so on. However, these methods which are based on virtual machine usually cause a large overhead. Thus, using those methods to estimate WSS is impractical. In this paper, we propose a novel framework to efficiently estimate WSS with eBPF (extended Berkeley Packet Filter), a cutting-edge technology which monitors and filters data by being attached to the kernel. With an eBPF program pinned into the kernel, we get the times of page fault and other information of memory allocation. Moreover, we collect WSS via vanilla tool to train a predictive model to complete estimation work with LightGBM, a useful tool which performs well on generating decision trees over continuous value. The experimental results illustrate that our framework can estimate WSS precisely with 98.5\% reduction in overhead compared to traditional methods.

Designing visually appealing layouts for multimedia documents containing text, graphs and images requires a form of creative intelligence. Modelling the generation of layouts has recently gained attention due to its importance in aesthetics and communication style. In contrast to standard prediction tasks, there are a range of acceptable layouts which depend on user preferences. For example, a poster designer may prefer logos on the top-left while another prefers logos on the bottom-right. Both are correct choices yet existing machine learning models treat layouts as a single choice prediction problem. In such situations, these models would simply average over all possible choices given the same input forming a degenerate sample. In the above example, this would form an unacceptable layout with a logo in the centre. In this paper, we present an auto-regressive neural network architecture, called LayoutMCL, that uses multi-choice prediction and winner-takes-all loss to effectively stabilise layout generation. LayoutMCL avoids the averaging problem by using multiple predictors to learn a range of possible options for each layout object. This enables LayoutMCL to generate multiple and diverse layouts from a single input which is in contrast with existing approaches which yield similar layouts with minor variations. Through quantitative benchmarks on real data (magazine, document and mobile app layouts), we demonstrate that LayoutMCL reduces Fr\'echet Inception Distance (FID) by 83-98% and generates significantly more diversity in comparison to existing approaches.