Affinity Posters
Tiny Papers Poster Session 5
Krystal Maughan · Thomas F Burns
Halle B
Live content is unavailable. Log in and register to view live content
Schedule
Thu 1:45 a.m. - 3:45 a.m.
|
Adverb Is the Key: Simple Text Data Augmentation with Adverb Deletion
(
Poster
#305
)
>
link
Poster Location: Halle B #305 In the field of text data augmentation, rule-based methods are widely adopted for real-world applications owing to their cost-efficiency. However, conventional rule-based approaches suffer from the possibility of losing the original semantics of the given text. We propose a novel text data augmentation strategy that avoids such phenomena through a straightforward deletion of adverbs, which play a subsidiary role in the sentence. Our comprehensive experiments demonstrate the efficiency and effectiveness of our proposed approach for not just single text classification, but also natural language inference that requires semantic preservation. We publicly released our source code for reproducibility. |
Juhwan Choi · Youngbin Kim 🔗 |
Thu 1:45 a.m. - 3:45 a.m.
|
Loss-Free Machine Unlearning
(
Poster
#304
)
>
link
Poster Location: Halle B #304 We present a machine unlearning approach that is both retraining- and label-free. Most existing machine unlearning approaches require a model to be fine-tuned to remove information while preserving performance. This is computationally expensive and necessitates the storage of the whole dataset for the lifetime of the model. Retraining-free approaches often utilise Fisher information, which is derived from the loss and requires labelled data which may not be available. Thus, we present an extension to the Selective Synaptic Dampening algorithm, substituting the diagonal of the Fisher information matrix for the gradient of the $l_2$ norm of the model output to approximate sensitivity. We evaluate our method in a range of experiments using ResNet18 and Vision Transformer. Results show our label-free method is competitive with existing state-of-the-art approaches. |
Jack Foster · Stefan Schoepf · Alexandra Brintrup 🔗 |
Thu 1:45 a.m. - 3:45 a.m.
|
Utilizing Cross-Version Consistency for Domain Adaptation: A Case Study on Music Audio
(
Poster
#303
)
>
link
Poster Location: Halle B #303 Deep learning models are commonly trained on large annotated corpora, often in a specific domain. Generalization to another domain without annotated data is usually challenging. In this paper, we address such unsupervised domain adaptation based on the teacher--student learning paradigm. For improved efficacy in the target domain, we propose to exploit cross-version scenarios, i.e., corresponding data pairs assumed to obtain the same yet unknown labels. More specifically, our idea is to compare teacher annotations across versions and use only consistent annotations as labels to train the student model. Examples of cross-version data include the same text by different speakers (in speech recognition) or the same character by different writers (in handwritten text recognition). In our case study on music audio, versions are different recorded performances of the same composition, aligned with music synchronization techniques. Taking pitch estimation (a multi-label classification task) as an example task, we show that enforcing consistency across versions in student training helps to improve the transfer from a source domain (piano) to unseen and more complex target domains (singing/orchestra). |
Lele Liu · Christof Weiß 🔗 |
Thu 1:45 a.m. - 3:45 a.m.
|
DFWLayer: Differentiable Frank-Wolfe Optimization Layer
(
Poster
#302
)
>
link
Poster Location: Halle B #302 Differentiable optimization has received a significant amount of attention due to its foundational role in the domain of machine learning based on neural networks. This paper proposes a differentiable layer, named Differentiable Frank-Wolfe Layer (DFWLayer), by rolling out the Frank-Wolfe method, a well-known optimization algorithm which can solve constrained optimization problems without projections and Hessian matrix computations, thus leading to an efficient way of dealing with large-scale convex optimization problems with norm constraints. Experimental results demonstrate that the DFWLayer not only attains competitive accuracy in solutions and gradients but also consistently adheres to constraints. |
Zixuan Liu · Liu Liu · Xueqian Wang · Peilin Zhao 🔗 |
Thu 1:45 a.m. - 3:45 a.m.
|
Revamp: Automated Simulations of Adversarial Attacks on Arbitrary Objects in Realistic Scenes
(
Poster
#301
)
>
link
Poster Location: Halle B #301 Deep learning models, such as those used in autonomous vehicles are vulnerable to adversarial attacks where attackers could place adversarial objects in the environment to induce incorrect detections. While generating such adversarial objects in the digital realm is well-studied, successfully transferring these attacks to the physical realm remains challenging, especially when accounting for real-world environmental factors. We address these challenges with REVAMP, a first-of-its-kind Python library for creating attack scenarios with arbitrary objects in scenes with realistic environmental factors, lighting, reflection, and refraction. REVAMP empowers researchers and practitioners to swiftly explore diverse scenarios, offering a wide range of configurable options for experiment design and using differentiable rendering to replicate physically-plausible adversarial objects. REVAMP is open-source and available at https://anonymous.4open.science/r/revamp and a demo video is available at https://youtu.be/ogCRO15R7-E. |
Matthew Hull · Zijie Wang · Polo Chau 🔗 |
Thu 1:45 a.m. - 3:45 a.m.
|
DSF-GAN: Downstream Feedback Generative Adversarial Network
(
Poster
#300
)
>
link
Poster Location: Halle B #300 Utility and privacy are two crucial measurements of synthetic tabular data. While privacy measures have been dramatically improved with the use of Generative Adversarial Networks (GANs), generating high-utility synthetic samples remains challenging. To increase the samples' utility, we propose a novel architecture called DownStream Feedback Generative Adversarial Network (DSF-GAN). This approach uses feedback from a downstream prediction model mid-training, to add valuable information to the generator’s loss function. Hence, DSF-GAN harnesses a downstream prediction task to increase the utility of the synthetic samples. To properly evaluate our method, we tested it using two popular data sets. Our experiments show better model performance when training on DSF-GAN-generated synthetic samples compared to synthetic data generated using the same GAN architecture without feedback when evaluated on the same validation set comprised of real samples. All code and datasets used in this research are openly available for ease of reproduction. |
Oriel Perets · Nadav Rappoport 🔗 |