Keeping Deep Learning Models in Check: A History-Based Approach to Mitigate Overfitting

In software engineering, deep learning models are increasingly deployed for critical tasks such as bug detection and code review. However, overfitting remains a challenge that affects the quality, reliability, and trustworthiness of software systems that utilize deep learning models. Overfitting can be (1) prevented (e.g., using dropout or early stopping) or (2) detected in a trained model (e.g., using correlation-based approaches). Both overfitting detection and prevention approaches that are currently used have constraints (e.g., requiring modification of the model structure, and high computing resources). In this paper, we propose a simple, yet powerful approach that can both detect and prevent overfitting based on the training history (i.e., validation losses). Our approach first trains a time series classifier on training histories of overfit models. This classifier is then used to detect if a trained model is overfit. In addition, our trained classifier can be used to prevent overfitting by identifying the optimal point to stop a model's training. We evaluate our approach on its ability to identify and prevent overfitting in real-world samples. We compare our approach against correlation-based detection approaches and the most commonly used prevention approach (i.e., early stopping). Our approach achieves an F1 score of 0.91 which is at least 5% higher than the current best-performing non-intrusive overfitting detection approach. Furthermore, our approach can stop training to avoid overfitting at least 32% of the times earlier than early stopping and has the same or a better rate of returning the best model.

GlitchBench: Can large multimodal models detect video game glitches?

Large multimodal models (LMMs) have evolved from large language models (LLMs) to integrate multiple input modalities, such as visual inputs. This integration augments the capacity of LLMs for tasks requiring visual comprehension and reasoning. However, the extent and limitations of their enhanced abilities are not fully understood, especially when it comes to real-world tasks. To address this gap, we introduce GlitchBench, a novel benchmark derived from video game quality assurance tasks, to test and evaluate the reasoning capabilities of LMMs. Our benchmark is curated from a variety of unusual and glitched scenarios from video games and aims to challenge both the visual and linguistic reasoning powers of LMMs in detecting and interpreting out-of-the-ordinary events. We evaluate multiple state-of-the-art LMMs, and we show that GlitchBench presents a new challenge for these models. Code and data are available at: https://glitchbench.github.io/

Zoom is what you need: An empirical study of the power of zoom and spatial biases in image classification

Image classifiers are information-discarding machines, by design. Yet, how these models discard information remains mysterious. We hypothesize that one way for image classifiers to reach high accuracy is to first zoom to the most discriminative region in the image and then extract features from there to predict image labels. We study six popular networks ranging from AlexNet to CLIP and find that proper framing of the input image can lead to the correct classification of 98.91% of ImageNet images. Furthermore, we explore the potential and limits of zoom transforms in image classification and uncover positional biases in various datasets, especially a strong center bias in two popular datasets: ImageNet-A and ObjectNet. Finally, leveraging our insights into the potential of zoom, we propose a state-of-the-art test-time augmentation (TTA) technique that improves classification accuracy by forcing models to explicitly perform zoom-in operations before making predictions. Our method is more interpretable, accurate, and faster than MEMO, a state-of-the-art TTA method. Additionally, we propose ImageNet-Hard, a new benchmark where zooming in alone often does not help state-of-the-art models better label images.

Large Language Models are Pretty Good Zero-Shot Video Game Bug Detectors

Video game testing requires game-specific knowledge as well as common sense reasoning about the events in the game. While AI-driven agents can satisfy the first requirement, it is not yet possible to meet the second requirement automatically. Therefore, video game testing often still relies on manual testing, and human testers are required to play the game thoroughly to detect bugs. As a result, it is challenging to fully automate game testing. In this study, we explore the possibility of leveraging the zero-shot capabilities of large language models for video game bug detection. By formulating the bug detection problem as a question-answering task, we show that large language models can identify which event is buggy in a sequence of textual descriptions of events from a game. To this end, we introduce the GameBugDescriptions benchmark dataset, which consists of 167 buggy gameplay videos and a total of 334 question-answer pairs across 8 games. We extensively evaluate the performance of six models across the OPT and InstructGPT large language model families on our benchmark dataset. Our results show promising results for employing language models to detect video game bugs. With the proper prompting technique, we could achieve an accuracy of 70.66%, and on some video games, up to 78.94%. Our code, evaluation data and the benchmark can be found on https://asgaardlab.github.io/LLMxBugs

CLIP meets GamePhysics: Towards bug identification in gameplay videos using zero-shot transfer learning

Gameplay videos contain rich information about how players interact with the game and how the game responds. Sharing gameplay videos on social media platforms, such as Reddit, has become a common practice for many players. Often, players will share gameplay videos that showcase video game bugs. Such gameplay videos are software artifacts that can be utilized for game testing, as they provide insight for bug analysis. Although large repositories of gameplay videos exist, parsing and mining them in an effective and structured fashion has still remained a big challenge. In this paper, we propose a search method that accepts any English text query as input to retrieve relevant videos from large repositories of gameplay videos. Our approach does not rely on any external information (such as video metadata); it works solely based on the content of the video. By leveraging the zero-shot transfer capabilities of the Contrastive Language-Image Pre-Training (CLIP) model, our approach does not require any data labeling or training. To evaluate our approach, we present the $\texttt{GamePhysics}$ dataset consisting of 26,954 videos from 1,873 games, that were collected from the GamePhysics section on the Reddit website. Our approach shows promising results in our extensive analysis of simple queries, compound queries, and bug queries, indicating that our approach is useful for object and event detection in gameplay videos. An example application of our approach is as a gameplay video search engine to aid in reproducing video game bugs. Please visit the following link for the code and the data: https://asgaardlab.github.io/CLIPxGamePhysics/

Studying Popular Open Source Machine Learning Libraries and Their Cross-Ecosystem Bindings

Open source machine learning (ML) libraries allow developers to integrate advanced ML functionality into their own applications. However, popular ML libraries, such as TensorFlow, are not available natively in all programming languages and software package ecosystems. Hence, developers who wish to use an ML library which is not available in their programming language or ecosystem of choice, may need to resort to using a so-called binding library. Binding libraries provide support across programming languages and package ecosystems for a source library. For example, the Keras .NET binding provides support for the Keras library in the NuGet (.NET) ecosystem even though the Keras library was written in Python. In this paper, we conduct an in-depth study of 155 cross-ecosystem bindings and their development for 36 popular open source ML libraries. Our study shows that for most popular ML libraries, only one package ecosystem is officially supported (usually PyPI). Cross-ecosystem support, which is available for 25% of the studied ML libraries, is usually provided through community-maintained bindings, e.g., 73% of the bindings in the npm ecosystem are community-maintained. Our study shows that the vast majority of the studied bindings cover only a small portion of the source library releases, and the delay for receiving support for a source library release is large.

Applications of Generative Adversarial Networks in Anomaly Detection: A Systematic Literature Review

Anomaly detection has become an indispensable tool for modern society, applied in a wide range of applications, from detecting fraudulent transactions to malignant brain tumours. Over time, many anomaly detection techniques have been introduced. However, in general, they all suffer from the same problem: a lack of data that represents anomalous behaviour. As anomalous behaviour is usually costly (or dangerous) for a system, it is difficult to gather enough data that represents such behaviour. This, in turn, makes it difficult to develop and evaluate anomaly detection techniques. Recently, generative adversarial networks (GANs) have attracted a great deal of attention in anomaly detection research, due to their unique ability to generate new data. In this paper, we present a systematic literature review of the applications of GANs in anomaly detection, covering 128 papers on the subject. The goal of this review paper is to analyze and summarize: (1) which anomaly detection techniques can benefit from certain types of GANs, and how, (2) in which application domains GAN-assisted anomaly detection techniques have been applied, and (3) which datasets and performance metrics have been used to evaluate these techniques. Our study helps researchers and practitioners to find the most suitable GAN-assisted anomaly detection technique for their application. In addition, we present a research roadmap for future studies in this area.