Characterizing Image Recognition Difficulty in Artificial and Biological Visual Processing by Cummings, Jesse E.

“…In recent years, computational models trained to do object recognition have become increasingly capable. …”
Get full text

Natural Language Control for for Visually Interactive Decision Support Tools in Supply Chain Management by Guter, Willem J.

“…Additionally, natural language interfaces can be difficult to implement, and require applications specific programming or training. This thesis proposes integrating a pre-trained large language model as the natural language interface. …”
Get full text

Image analysis for building facade inspection by Fan, Yaxi

“…By developing the automated system in the façade inspection process is expected to reduce risks to trained personnel who is climbing up to scaffolding structures and other supporting equipment for diagnostics work. …”
Get full text

AI in urban planning by Low, Ryan Wai Zhun

“…Machine learning model is trained with this dataset to accurately predict and rank candidate locations based on their google ratings. …”
Get full text

Generalized AutoNLP model for name entity recognition task by Wong, Yung Shen

“…Unsupervised pre-trained word embeddings have been widely used in recent studies in the field of Natural Language Processing. …”
Get full text

Unsupervised domain adaptation for depth completion from sparse LiDAR scans depth map by Geng, Yue

“…Denser scans depth input leads to better prediction, while the cost of the corresponding LiDAR equipment will be more expensive, and the model trained by dense depth input performs badly on sparse depth input. …”
Get full text

Embedded system application development on Raspberry Pi 4 (alpha crop weed and disease detection) by Goh, Jun De

“…The proposed system utilises deep learning and machine learning algorithms to analyse images of cotton plants captured by a camera attached to the Raspberry Pi. The system is trained using a custom dataset to recognize disease and weed patterns in cotton plants. …”
Get full text

Machine learning-based local collision avoidance for maritime navigation by Zou, Yixuan

“…As a former study for the model training in sea environment, agent was trained in various ground environment using TD3 algorithm during the research. …”
Get full text

Llama2 self-improvement using memory-of-thought by Dong, Yuxiu

“…This project evaluates the effectiveness of MoT on a pre-trained Large Language Model Llama2 and compares the performance of improved Llama2 with ChatGPT3.5 API on 10 benchmark data sets. …”
Get full text

Characterization of hippocampal CA1 place cells. by Lau, Hwee Hui.

“…In this study, rats were pre-trained in a square box prior to surgical implantation of microdrives into the brain for extracellular recording in the hippocampal CA1 pyramidal layer. …”
Get full text

Visual search and application using deep learning (Age group classification with convolution neural network) by Low, Benjamin

“…In this paper, a deep CNN model that was trained for face recognition task is used to estimate the age information on the IMDB-WIKI database. …”
Get full text

As firm as their foundations: creating transferable adversarial examples across downstream tasks with CLIP by Hu, A, Gu, J, Pinto, F, Kamnitsas, K, Torr, PHS

“…Foundation models pre-trained on web-scale vision-language data, such as CLIP, are widely used as cornerstones of powerful machine learning systems. …”

Large scale paired antibody language models by Kenlay, H, Dreyer, FA, Kovaltsuk, A, Miketa, D, Pires, D, Deane, CM

“…To address this challenge, we present IgBert and IgT5, the best performing antibody-specific language models developed to date which can consistently handle both paired and unpaired variable region sequences as input. These models are trained comprehensively using the more than two billion unpaired sequences and two million paired sequences of light and heavy chains present in the Observed Antibody Space dataset. …”

Incremental learning of object detectors using a visual shape alphabet by Opelt, A, Pinz, A, Zisserman, A

“…We show that the sharing of shape features not only reduces the number of features required per category, but also often improves recognition performance, as compared to individual detectors which are trained on a per-class basis.…”

wav2sleep: a unified multi-modal approach to sleep stage classification from physiological signals by Carter, JF, Tarassenko, L

“…Existing approaches to this problem have typically used deep learning models designed and trained to operate on one or more specific input signals. …”

Interpreting learned feedback patterns in large language models by Marks, L, Abdullah, A, Neo, C, Arike, R, Krueger, D, Torr, P, Barez, F

“…Reinforcement learning from human feedback (RLHF) is widely used to train large language models (LLMs). However, it is unclear whether LLMs accurately learn the underlying preferences in human feedback data. …”

Community noise measure via crowd sourcing by Lim, Yun Jie

“…Since GVI only quantifies the presence of trees instead of its actual location, the experiment also looks at using TensorFlow Object Identification API to build a tree detection algorithm. Various pre-trained models obtained from TensorFlow 2 Detection Model Zoo were trained and tested and the Single-shot MultiBox Detector (SSD) with MobileNet was observed to perform the best. …”
Get full text

A promising deep learning-assistive algorithm for histopathological screening of colorectal cancer by Ho, Cowan, Zhao, Zitong, Chen, Xiu Fen, Sauer, Jan, Saraf, Sahil Ajit, Jialdasani, Rajasa, Taghipour, Kaveh, Sathe, Aneesh, Khor, Li-Yan, Lim, Kiat-Hon, Leow, Wei-Qiang

“…With the increasing number of colonoscopies being performed, colorectal biopsies make up a large proportion of any histopathology laboratory workload. We trained and validated a unique artificial intelligence (AI) deep learning model as an assistive tool to screen for colonic malignancies in colorectal specimens, in order to improve cancer detection and classification; enabling busy pathologists to focus on higher order decision-making tasks. …”
Get full text

Evolving code with a large language model by Hemberg, Erik, Moskal, Stephen, O’Reilly, Una-May

“…Like GP, it uses evolutionary operators, but its designs and implementations of those operators significantly differ from GP’s because they enlist an LLM, using prompting and the LLM’s pre-trained pattern matching and sequence completion capability. …”
Get full text

Investigating Fine-Tuning of Language Models for Multiple-Choice Questions by Wang, Ivy A.

“…We specifically investigate training data properties related to positional bias in fine-tuned language model performance on correctly answering MCQs. …”
Get full text