Author
Andy Shih
Other affiliations: Stanford University
Bio: Andy Shih is an academic researcher from University of California, Los Angeles. The author has contributed to research in topics: Probabilistic logic & Graphical model. The author has an hindex of 8, co-authored 17 publications receiving 253 citations. Previous affiliations of Andy Shih include Stanford University.
Papers
More filters
•
TL;DR: The authors provides a thorough account of the opportunities and risks of foundation models, ranging from their capabilities (e.g., language, vision, robotics, reasoning, human interaction) and technical principles(e. g.g. model architectures, training procedures, data, systems, security, evaluation, theory) to their applications.
Abstract: AI is undergoing a paradigm shift with the rise of models (e.g., BERT, DALL-E, GPT-3) that are trained on broad data at scale and are adaptable to a wide range of downstream tasks. We call these models foundation models to underscore their critically central yet incomplete character. This report provides a thorough account of the opportunities and risks of foundation models, ranging from their capabilities (e.g., language, vision, robotics, reasoning, human interaction) and technical principles(e.g., model architectures, training procedures, data, systems, security, evaluation, theory) to their applications (e.g., law, healthcare, education) and societal impact (e.g., inequity, misuse, economic and environmental impact, legal and ethical considerations). Though foundation models are based on standard deep learning and transfer learning, their scale results in new emergent capabilities,and their effectiveness across so many tasks incentivizes homogenization. Homogenization provides powerful leverage but demands caution, as the defects of the foundation model are inherited by all the adapted models downstream. Despite the impending widespread deployment of foundation models, we currently lack a clear understanding of how they work, when they fail, and what they are even capable of due to their emergent properties. To tackle these questions, we believe much of the critical research on foundation models will require deep interdisciplinary collaboration commensurate with their fundamentally sociotechnical nature.
76 citations
••
01 Jul 2018TL;DR: In this article, the authors propose an approach for explaining Bayesian network classifiers, which is based on compiling such classifiers into decision functions that have a tractable and symbolic form.
Abstract: We propose an approach for explaining Bayesian network classifiers, which is based on compiling such classifiers into decision functions that have a tractable and symbolic form. We introduce two types of explanations for why a classifier may have classified an instance positively or negatively and suggest algorithms for computing these explanations. The first type of explanation identifies a minimal set of the currently active features that is responsible for the current classification, while the second type of explanation identifies a minimal set of features whose current state (active or not) is sufficient for the classification. We consider in particular the compilation of Naive and Latent-Tree Bayesian network classifiers into Ordered Decision Diagrams (ODDs), providing a context for evaluating our proposal using case studies and experiments based on classifiers from the literature.
64 citations
••
17 Jul 2019TL;DR: An algorithm is proposed for compiling Bayesian network classifiers into decision graphs that mimic the input and output behavior of the classifiers, which are tractable and can be exponentially smaller in size than decision trees.
Abstract: We propose an algorithm for compiling Bayesian network classifiers into decision graphs that mimic the input and output behavior of the classifiers. In particular, we compile Bayesian network classifiers into ordered decision graphs, which are tractable and can be exponentially smaller in size than decision trees. This tractability facilitates reasoning about the behavior of Bayesian network classifiers, including the explanation of decisions they make. Our compilation algorithm comes with guarantees on the time of compilation and the size of compiled decision graphs. We apply our compilation algorithm to classifiers from the literature and discuss some case studies in which we show how to automatically explain their decisions and verify properties of their behavior.
49 citations
••
07 Jul 2019TL;DR: An Angluin-style learning algorithm is proposed to compile a neural network on a given region into an Ordered Binary Decision Diagram (OBDD), using a SAT solver as an equivalence oracle to verify the behavior of binarized neural networks.
Abstract: We consider the problem of verifying the behavior of binarized neural networks on some input region. We propose an Angluin-style learning algorithm to compile a neural network on a given region into an Ordered Binary Decision Diagram (OBDD), using a SAT solver as an equivalence oracle. The OBDD allows us to efficiently answer a range of verification queries, including counting, computing the probability of counterexamples, and identifying common characteristics of counterexamples. We also present experimental results on verifying binarized neural networks that recognize images of handwritten digits.
43 citations
••
09 Jul 2020TL;DR: A more efficient approach for compiling neural networks is considered, based on a pseudo-polynomial time algorithm for compiling a neuron, and it is shown that it is feasible to obtain compact representations of neural networks as SDDs.
Abstract: We consider the compilation of a binary neural network's decision function into tractable representations such as Ordered Binary Decision Diagrams (OBDDs) and Sentential Decision Diagrams (SDDs). Obtaining this function as an OBDD/SDD facilitates the explanation and formal verification of a neural network's behavior. First, we consider the task of verifying the robustness of a neural network, and show how we can compute the expected robustness of a neural network, given an OBDD/SDD representation of it. Next, we consider a more efficient approach for compiling neural networks, based on a pseudo-polynomial time algorithm for compiling a neuron. We then provide a case study in a handwritten digits dataset, highlighting how two neural networks trained from the same dataset can have very high accuracies, yet have very different levels of robustness. Finally, in experiments, we show that it is feasible to obtain compact representations of neural networks as SDDs.
41 citations
Cited by
More filters
•
92 citations
•
06 Sep 2019TL;DR: It is demonstrated that explanations and adversarial examples are related by a generalized form of hitting set duality, which extends earlier work on hitting setDuality observed in model-based diagnosis and knowledge compilation.
Abstract: The importance of explanations (XP's) of machine learning (ML) model predictions and of adversarial examples (AE's) cannot be overstated, with both arguably being essential for the practical success of ML in different settings. There has been recent work on understanding and assessing the relationship between XP's and AE's. However, such work has been mostly experimental and a sound theoretical relationship has been elusive. This paper demonstrates that explanations and adversarial examples are related by a generalized form of hitting set duality, which extends earlier work on hitting set duality observed in model-based diagnosis and knowledge compilation. Furthermore, the paper proposes algorithms, which enable computing adversarial examples from explanations and vice-versa.
84 citations
•
TL;DR: A theory for unveiling the reasons behind the decisions made by Boolean classifiers is presented and notions such as sufficient, necessary and complete reasons behind decisions are defined, in addition to classifier and decision bias.
Abstract: Recent work has shown that some common machine learning classifiers can be compiled into Boolean circuits that have the same input-output behavior. We present a theory for unveiling the reasons behind the decisions made by Boolean classifiers and study some of its theoretical and practical implications. We define notions such as sufficient, necessary and complete reasons behind decisions, in addition to classifier and decision bias. We show how these notions can be used to evaluate counterfactual statements such as "a decision will stick even if ... because ... ." We present efficient algorithms for computing these notions, which are based on new advances on tractable Boolean circuits, and illustrate them using a case study.
66 citations