Reinforcement learning is becoming ever more prominent in solving combinatorial search problems, in particular ones where states are images. Prior work has devised action-policy testing methodology, that identifies so-called bug states where policy performance is sub-optimal. Here we show how to leverage this methodology during the RL process, using action-policy testing to find bugs and injecting those as alternate start states for the training runs. Running experiments across six 2D games, we find that our testing-guided training often achieves similar expected reward while reducing the number of bugs.

Large language models are becoming increasingly practical for translating code across programming languages, a process known as transpiling. Even though automated transpilation significantly boosts developer productivity, a key concern is whether the generated code is correct. Existing work initially used manually crafted test suites to test the translations of a small corpus of programs; these test suites were later automated. In contrast, we devise the first approach for automated, functional, property-based testing of code translation models. Our general, user-provided specifications about the transpiled code capture a range of properties, from purely syntactic to purely semantic ones. As shown by our experiments, this approach is very effective in detecting property violations in popular code translation models, and therefore, in evaluating model quality with respect to given properties. We also go a step further and explore the usage scenario where a user simply aims to obtain a correct translation of some code with respect to certain properties without necessarily being concerned about the overall quality of the model. To this purpose, we develop the first property-guided search procedure for code translation models, where a model is repeatedly queried with slightly different parameters to produce alternative and potentially more correct translations. Our results show that this search procedure helps to obtain significantly better code translations.

Machine-learning models are becoming increasingly prevalent in our lives, for instance assisting in image-classification or decision-making tasks. Consequently, the reliability of these models is of critical importance and has resulted in the development of numerous approaches for validating and verifying their robustness and fairness. However, beyond such specific properties, it is challenging to specify, let alone check, general functional-correctness expectations from models. In this paper, we take inspiration from specifications used in formal methods, expressing functional-correctness properties by reasoning about k different executions---so-called k-safety properties. Considering a credit-screening model of a bank, the expected property that "if a person is denied a loan and their income decreases, they should still be denied the loan" is a 2-safety property. Here, we show the wide applicability of k-safety properties for machine-learning models and present the first specification language for expressing them. We also operationalize the language in a framework for automatically validating such properties using metamorphic testing. Our experiments show that our framework is effective in identifying property violations, and that detected bugs could be used to train better models.

Testing is a promising way to gain trust in a learned action policy π, in particular if π is a neural network. A “bug” in this context constitutes undesirable or fatal policy behavior, e.g., satisfying a failure condition. But how do we distinguish whether such behavior is due to bad policy decisions, or whether it is actually unavoidable under the given circumstances? This requires knowledge about optimal solutions, which defeats the scalability of testing. Related problems occur in software testing when the correct program output is not known.

Abstract State-of-the-art program-analysis techniques are not yet able to effectively verify safety properties of heterogeneous systems, that is, systems with components implemented using diverse technologies. This shortcoming is pinpointed by programs invoking neural networks despite their acclaimed role as innovation drivers across many application areas. In this paper, we embark on the verification of system-level properties for systems characterized by interaction between programs and neural networks. Our technique provides a tight two-way integration of a program and a neural-network analysis and is formalized in a general framework based on abstract interpretation. We evaluate its effectiveness on 26 variants of a widely used, restricted autonomous-driving benchmark.

Deep Learning (DL) systems are key enablers for engineering intelligent applications due to their ability to solve complex tasks such as image recognition and machine translation. Nevertheless, using DL systems in safety- and security-critical applications requires to provide testing evidence for their dependable operation. Recent research in this direction focuses on adapting testing criteria from traditional software engineering as a means of increasing confidence for their correct behaviour. However, they are inadequate in capturing the intrinsic properties exhibited by these systems. We bridge this gap by introducing DeepImportance, a systematic testing methodology accompanied by an Importance-Driven (IDC) test adequacy criterion for DL systems. Applying IDC enables to establish a layer-wise functional understanding of the importance of DL system components and use this information to assess the semantic diversity of a test set. Our empirical evaluation on several DL systems, across multiple DL datasets and with state-of-the-art adversarial generation techniques demonstrates the usefulness and effectiveness of DeepImportance and its ability to support the engineering of more robust DL systems.

Deep Neural Networks (DNNs) are increasingly deployed in safety-critical applications including autonomous vehicles and medical diagnostics. To reduce the residual risk for unexpected DNN behaviour and provide evidence for their trustworthy operation, DNNs should be thoroughly tested. The DeepFault whitebox DNN testing approach presented in our paper addresses this challenge by employing suspiciousness measures inspired by fault localization to establish the hit spectrum of neurons and identify suspicious neurons whose weights have not been calibrated correctly and thus are considered responsible for inadequate DNN performance. DeepFault also uses a suspiciousness-guided algorithm to synthesize new inputs, from correctly classified inputs, that increase the activation values of suspicious neurons. Our empirical evaluation on several DNN instances trained on MNIST and CIFAR-10 datasets shows that DeepFault is effective in identifying suspicious neurons. Also, the inputs synthesized by DeepFault closely resemble the original inputs, exercise the identified suspicious neurons and are highly adversarial.

We consider a case study of the problem of deploying an autonomous air vehicle in a partially observable, dynamic, indoor environment from a specification given as a linear temporal logic (LTL) formula over regions of interest. We model the motion and sensing capabilities of the vehicle as a partially observable Markov decision process (POMDP). We adapt recent results for solving POMDPs with parity objectives to generate a control policy. We also extend the existing framework with a policy minimization technique to obtain a better implementable policy, while preserving its correctness. The proposed techniques are illustrated in an experimental setup involving an autonomous quadrotor performing surveillance in a dynamic environment.