no code implementations • 4 Mar 2024 • Saeed Najafi, Alona Fyshe
Pre-trained Language Models (PLMs) can be accurately fine-tuned for downstream text processing tasks.
no code implementations • 7 Jan 2024 • Greta Tuckute, Dawn Finzi, Eshed Margalit, Joel Zylberberg, SueYeon Chung, Alona Fyshe, Evelina Fedorenko, Nikolaus Kriegeskorte, Jacob Yates, Kalanit Grill Spector, Kohitij Kar
In recent years, neuroscience has made significant progress in building large-scale artificial neural network (ANN) models of brain activity and behavior.
no code implementations • 26 Oct 2023 • Daniela Teodorescu, Tiffany Cheng, Alona Fyshe, Saif M. Mohammad
Research in psychopathology has shown that, at an aggregate level, the patterns of emotional change over time -- emotion dynamics -- are indicators of one's mental health.
no code implementations • 11 Jul 2023 • Dhruv Mullick, Bilal Ghanem, Alona Fyshe
Customer feedback is invaluable to companies as they refine their products.
no code implementations • 8 Jun 2023 • Daniela Teodorescu, Alona Fyshe, Saif M. Mohammad
These results act as a useful baselines for further research in how patterns of emotions expressed by children change with age, and their association with mental health.
no code implementations • 6 Jun 2023 • Cory Efird, Alex Murphy, Joel Zylberberg, Alona Fyshe
In the final section of our analysis, we localize shared decodable concepts in the brain using a voxel-masking optimization method to produce a shared decodable concept (SDC) space.
no code implementations • 10 Apr 2023 • Bilal Ghanem, Alona Fyshe
At the same time, DISTO ranks the performance of state-of-the-art DG models very differently from MT-based metrics, showing that MT metrics should not be used for distractor evaluation.
1 code implementation • 21 Jan 2023 • Saeed Najafi, Alona Fyshe
Zero-Shot Relation Extraction (ZRE) is the task of Relation Extraction where the training and test sets have no shared relation types.
no code implementations • 6 Sep 2022 • Cassidy Pirlot, Richard C. Gerum, Cory Efird, Joel Zylberberg, Alona Fyshe
As convolutional neural networks (CNNs) become more accurate at object recognition, their representations become more similar to the primate visual system.
1 code implementation • 22 Aug 2022 • Richard C. Gerum, Cassidy Pirlot, Alona Fyshe, Joel Zylberberg
For convolutional networks, the best $\alpha$ values depend on the task complexity and evaluation metric: lower $\alpha$ values optimized validation accuracy and robustness to adversarial attack for networks performing a simple object recognition task (categorizing MNIST images of handwritten digits); for a more complex task (categorizing CIFAR-10 natural images), we found that lower $\alpha$ values optimized validation accuracy whereas higher $\alpha$ values optimized adversarial robustness.
no code implementations • 12 Jul 2022 • Liam Peet-Pare, Nidhi Hegde, Alona Fyshe
Fairness researchers in machine learning (ML) have coalesced around several fairness criteria which provide formal definitions of what it means for an ML model to be fair.
no code implementations • 6 Jun 2022 • Dhruv Mullick, Alona Fyshe, Bilal Ghanem
Aspect-based Sentiment Analysis (ABSA) helps to explain customers' opinions towards products and services.
Aspect-Based Sentiment Analysis Aspect-Based Sentiment Analysis (ABSA)
no code implementations • 18 May 2022 • Han Wang, Archit Sakhadeo, Adam White, James Bell, Vincent Liu, Xutong Zhao, Puer Liu, Tadashi Kozuno, Alona Fyshe, Martha White
The performance of reinforcement learning (RL) agents is sensitive to the choice of hyperparameters.
no code implementations • Findings (ACL) 2022 • Bilal Ghanem, Lauren Lutz Coleman, Julia Rivard Dexter, Spencer McIntosh von der Ohe, Alona Fyshe
We show that the HTA-WTA model tests for strong SCRS by asking deep inferential questions.
no code implementations • ICLR 2022 • Kirby Banman, Liam Peet-Pare, Nidhi Hegde, Alona Fyshe, Martha White
In this work, we show that SGDm under covariate shift with a fixed step-size can be unstable and diverge.
no code implementations • 29 May 2021 • Qingfeng Lan, Luke Kumar, Martha White, Alona Fyshe
Correlates of secondary information appear in LSTM representations even though they are not part of an \emph{explicitly} supervised prediction task.
no code implementations • Findings of the Association for Computational Linguistics 2020 • Maryam Hashemzadeh, Greta Kaufeld, Martha White, Andrea E. Martin, Alona Fyshe
The representations generated by many models of language (word embeddings, recurrent neural networks and transformers) correlate to brain activity recorded while people read.
1 code implementation • ICLR 2020 • Qingfeng Lan, Yangchen Pan, Alona Fyshe, Martha White
Q-learning suffers from overestimation bias, because it approximates the maximum action value using the maximum estimated action value.
1 code implementation • 25 May 2019 • Callie Federer, Haoyan Xu, Alona Fyshe, Joel Zylberberg
To test this, we trained DCNNs on a composite task, wherein networks were trained to: a) classify images of objects; while b) having intermediate representations that resemble those observed in neural recordings from monkey visual cortex.
no code implementations • WS 2018 • Avery Hiebert, Cole Peterson, Alona Fyshe, Nishant Mehta
While Long Short-Term Memory networks (LSTMs) and other forms of recurrent neural network have been successfully applied to language modeling on a character level, the hidden state dynamics of these models can be difficult to interpret.
no code implementations • 27 Sep 2018 • Dhanush Dharmaretnam, Chris Foster, Alona Fyshe
Here we build on previous work that indicated that two very different model classes trained on two very different tasks actually build knowledge representations that have similar underlying representations.
no code implementations • WS 2018 • Sophia Chan, Alona Fyshe
Social media text is replete with unusual capitalization patterns.
no code implementations • NAACL 2018 • Dhanush Dharmaretnam, Alona Fyshe
Word vector models learn about semantics through corpora.
no code implementations • WS 2017 • Sophia Chan, Maryam Honari Jahromi, Benjamin Benetti, Aazim Lakhani, Alona Fyshe
Our team{---}Uvic-NLP{---}explored and evaluated a variety of lexical features for Native Language Identification (NLI) within the framework of ensemble methods.