Lendi Group is Australia's fastest-growing Fintech business and we're building a technology-enabled platform to take the hard work out of home loans. We're passionate about how technology can revolutionise our industry and solve a key pain point in peoples' lives. In an ever-changing regulatory environment, Lendi Group is leading the charge in developing industry-first technology designed to offer Australian home owners transparency, simplicity and convenience in their home loan experience. Data is forefront in this charge as we look to automate and build out scalable and reliable systems to support our customers and counterparty banks. Join of a growing team of 11 data professionals and use your inquisitive nature and attention to detail to surface key trends to the wider business.

In-game toxic language becomes the hot potato in the gaming industry and community. There have been several online game toxicity analysis frameworks and models proposed. However, it is still challenging to detect toxicity due to the nature of in-game chat, which has extremely short length. In this paper, we describe how the in-game toxic language shared task has been established using the real-world in-game chat data. In addition, we propose and introduce the model/framework for toxic language token tagging (slot filling) from the in-game chat. The data and code will be released.

Due to the noises in crowdsourced labels, label aggregation (LA) has emerged as a standard procedure to post-process crowdsourced labels. LA methods estimate true labels from crowdsourced labels by modeling worker qualities. Most existing LA methods are iterative in nature. They need to traverse all the crowdsourced labels multiple times in order to jointly and iteratively update true labels and worker qualities until convergence. Consequently, these methods have high space and time complexities. In this paper, we treat LA as a dynamic system and model it as a Dynamic Bayesian network. From the dynamic model we derive two light-weight algorithms, LA\textsuperscript{onepass} and LA\textsuperscript{twopass}, which can effectively and efficiently estimate worker qualities and true labels by traversing all the labels at most twice. Due to the dynamic nature, the proposed algorithms can also estimate true labels online without re-visiting historical data. We theoretically prove the convergence property of the proposed algorithms, and bound the error of estimated worker qualities. We also analyze the space and time complexities of the proposed algorithms and show that they are equivalent to those of majority voting. Experiments conducted on 20 real-world datasets demonstrate that the proposed algorithms can effectively and efficiently aggregate labels in both offline and online settings even if they traverse all the labels at most twice.

Intelligent human inputs are required both in the training and operation of AI systems, and within the governance of blockchain systems and decentralized autonomous organizations (DAOs). This paper presents a formal definition of Human Intelligence Primitives (HIPs), and describes the design and implementation of an Ethereum protocol for their on-chain collection, modeling, and integration in machine learning workflows.

This paper evaluates whether training a decision tree based on concepts extracted from a concept-based explainer can increase interpretability for Convolutional Neural Networks (CNNs) models and boost the fidelity and performance of the used explainer. CNNs for computer vision have shown exceptional performance in critical industries. However, it is a significant barrier when deploying CNNs due to their complexity and lack of interpretability. Recent studies to explain computer vision models have shifted from extracting low-level features (pixel-based explanations) to mid-or high-level features (concept-based explanations). The current research direction tends to use extracted features in developing approximation algorithms such as linear or decision tree models to interpret an original model. In this work, we modify one of the state-of-the-art concept-based explanations and propose an alternative framework named TreeICE. We design a systematic evaluation based on the requirements of fidelity (approximate models to original model's labels), performance (approximate models to ground-truth labels), and interpretability (meaningful of approximate models to humans). We conduct computational evaluation (for fidelity and performance) and human subject experiments (for interpretability) We find that Tree-ICE outperforms the baseline in interpretability and generates more human readable explanations in the form of a semantic tree structure. This work features how important to have more understandable explanations when interpretability is crucial.

Randomly masking and predicting word tokens has been a successful approach in pre-training language models for a variety of downstream tasks. In this work, we observe that the same idea also applies naturally to sequential decision making, where many well-studied tasks like behavior cloning, offline reinforcement learning, inverse dynamics, and waypoint conditioning correspond to different sequence maskings over a sequence of states, actions, and returns. We introduce the Uni[MASK] framework, which provides a unified way to specify models which can be trained on many different sequential decision making tasks. We show that a single Uni[MASK] model is often capable of carrying out many tasks with performance similar to or better than single-task models. Additionally, after fine-tuning, our Uni[MASK] models consistently outperform comparable single-task models. Our code is publicly available here.

A method for estimating the conditional average treatment effect under condition of censored time-to-event data called BENK (the Beran Estimator with Neural Kernels) is proposed. The main idea behind the method is to apply the Beran estimator for estimating the survival functions of controls and treatments. Instead of typical kernel functions in the Beran estimator, it is proposed to implement kernels in the form of neural networks of a specific form called the neural kernels. The conditional average treatment effect is estimated by using the survival functions as outcomes of the control and treatment neural networks which consists of a set of neural kernels with shared parameters. The neural kernels are more flexible and can accurately model a complex location structure of feature vectors. Various numerical simulation experiments illustrate BENK and compare it with the well-known T-learner, S-learner and X-learner for several types of the control and treatment outcome functions based on the Cox models, the random survival forest and the Nadaraya-Watson regression with Gaussian kernels. The code of proposed algorithms implementing BENK is available in https://github.com/Stasychbr/BENK.

Muscle tissue drives nearly all movement in the animal kingdom, providing power, mobility, and dexterity. Technologies for measuring muscle tissue motion, such as sonomicrometry, fluoromicrometry, and ultrasound, have significantly advanced our understanding of biomechanics. Yet, the field lacks the ability to monitor muscle tissue motion for animal behavior outside the lab. Towards addressing this issue, we previously introduced magnetomicrometry, a method that uses magnetic beads to wirelessly monitor muscle tissue length changes, and we validated magnetomicrometry via tightly-controlled in situ testing. In this study we validate the accuracy of magnetomicrometry against fluoromicrometry during untethered running in an in vivo turkey model. We demonstrate real-time muscle tissue length tracking of the freely-moving turkeys executing various motor activities, including ramp ascent and descent, vertical ascent and descent, and free roaming movement. Given the demonstrated capacity of magnetomicrometry to track muscle movement in untethered animals, we feel that this technique will enable new scientific explorations and an improved understanding of muscle function. -- -- El tejido muscular es el motor de casi todos los movimientos del reino animal, ya que proporciona fuerza, movilidad y destreza. Las tecnolog\'ias para medir el movimiento del tejido muscular, como la sonomicrometr\'ia, la fluoromicrometr\'ia y el ultrasonido, han avanzado considerablemente la comprensi\'on de la biomec\'anica. Sin embargo, este campo carece de la capacidad de rastrear el movimiento del tejido muscular en el comportamiento animal fuera del laboratorio. Para abordar este problema, presentamos previamente la magnetomicrometr\'ia, un m\'etodo que utiliza peque\~nos imanes para rastrear de forma inal\'ambrica los cambios de longitud del tejido muscular, y validamos la magnetomicrometr\'ia mediante pruebas estrechamente controladas in situ. En este estudio validamos la precisi\'on de la magnetomicrometr\'ia en comparaci\'on con la fluoromicrometr\'ia usando un modelo de pavo in vivo mientras corre libremente. Demostramos el rastreo en tiempo real de la longitud del tejido muscular de los pavos que se mueven libremente ejecutando varias actividades motoras, incluyendo el ascenso y el descenso en rampa, el ascenso y el descenso vertical, y el movimiento libre. Dada la capacidad demostrada de la magnetomicrometr\'ia para rastrear el movimiento muscular en animales en un contexto m\'ovil, creemos que esta t\'ecnica permitir\'a nuevas exploraciones cient\'ificas y una mejor comprensi\'on de la funci\'on muscular.

Large language models (LLMs) transfer well to new tasks out-of-the-box simply given a natural language prompt that demonstrates how to perform the task and no additional training. Prompting is a brittle process wherein small modifications to the prompt can cause large variations in the model predictions, and therefore significant effort is dedicated towards designing a painstakingly "perfect prompt" for a task. To mitigate the high degree of effort involved in prompt-design, we instead ask whether producing multiple effective, yet imperfect, prompts and aggregating them can lead to a high quality prompting strategy. Our observations motivate our proposed prompting method, ASK ME ANYTHING (AMA). We first develop an understanding of the effective prompt formats, finding that question-answering (QA) prompts, which encourage open-ended generation ("Who went to the park?") tend to outperform those that restrict the model outputs ("John went to the park. Output True or False."). Our approach recursively uses the LLM itself to transform task inputs to the effective QA format. We apply the collected prompts to obtain several noisy votes for the input's true label. We find that the prompts can have very different accuracies and complex dependencies and thus propose to use weak supervision, a procedure for combining the noisy predictions, to produce the final predictions for the inputs. We evaluate AMA across open-source model families (e.g., EleutherAI, BLOOM, OPT, and T0) and model sizes (125M-175B parameters), demonstrating an average performance lift of 10.2% over the few-shot baseline. This simple strategy enables the open-source GPT-J-6B model to match and exceed the performance of few-shot GPT3-175B on 15 of 20 popular benchmarks. Averaged across these tasks, the GPT-J-6B model outperforms few-shot GPT3-175B. We release our code here: https://github.com/HazyResearch/ama_prompting

Artificial intelligence (AI) systems attempt to imitate human behavior. How well they do this imitation is often used to assess their utility and to attribute human-like (or artificial) intelligence to them. However, most work on AI refers to and relies on human intelligence without accounting for the fact that human behavior is inherently shaped by the cultural contexts they are embedded in, the values and beliefs they hold, and the social practices they follow. Additionally, since AI technologies are mostly conceived and developed in just a handful of countries, they embed the cultural values and practices of these countries. Similarly, the data that is used to train the models also fails to equitably represent global cultural diversity. Problems therefore arise when these technologies interact with globally diverse societies and cultures, with different values and interpretive practices. In this position paper, we describe a set of cultural dependencies and incongruencies in the context of AI-based language and vision technologies, and reflect on the possibilities of and potential strategies towards addressing these incongruencies.