Matching Items (64)
Filtering by
- All Subjects: Computer Science
- Creators: Baral, Chitta
Description
In the era of information explosion and multi-modal data, information retrieval (IR) and question answering (QA) systems have become essential in daily human activities. IR systems aim to find relevant information in response to user queries, while QA systems provide concise and accurate answers to user questions. IR and QA are two of the most crucial challenges in the realm of Artificial Intelligence (AI), with wide-ranging real-world applications such as search engines and dialogue systems.
This dissertation investigates and develops novel models and training objectives to enhance current retrieval systems in textual and multi-modal contexts. Moreover, it examines QA systems, emphasizing generalization and robustness, and creates new benchmarks to promote their progress. Neural retrievers have surfaced as a viable solution, capable of surpassing the constraints of traditional term-matching search algorithms. This dissertation presents Poly-DPR, an innovative multi-vector model architecture that manages test-query, and ReViz, a comprehensive multimodal model to tackle multi-modality queries. By utilizing IR-focused pretraining tasks and producing large-scale training data, the proposed methodology substantially improves the abilities of existing neural retrievers.Concurrently, this dissertation investigates the realm of QA systems, referred to as ``readers'', by performing an exhaustive analysis of current extractive and generative readers, which results in a reliable guidance for selecting readers for downstream applications. Additionally, an original reader (Two-in-One) is designed to effectively choose the pertinent passages and sentences from a pool of candidates for multi-hop reasoning. This dissertation also acknowledges the significance of logical reasoning in real-world applications and has developed a comprehensive testbed, LogiGLUE, to further the advancement of reasoning capabilities in QA systems.
ContributorsLuo, Man (Author) / Baral, Chitta (Thesis advisor) / Yang, Yezhou (Committee member) / Blanco, Eduardo (Committee member) / Chen, Danqi (Committee member) / Arizona State University (Publisher)
Created2023
Description
Learning longer-horizon tasks is challenging with techniques such as reinforcement learning and behavior cloning. Previous approaches have split these long tasks into shorter tasks that are easier to learn by using statistical change point detection methods. However, classical changepoint detection methods function only with low-dimensional robot trajectory data and not with high-dimensional inputs such as vision. In this thesis, I have split a long horizon tasks, represented by trajectories into short-horizon sub-tasks with the supervision of language. These shorter horizon tasks can be learned using conventional behavior cloning approaches. I found comparisons between the techniques from the video moment retrieval problem and changepoint detection in robot trajectory data consisting of high-dimensional data. The proposed moment retrieval-based approach shows a more than 30% improvement in mean average precision (mAP) for identifying trajectory sub-tasks with language guidance compared to that without language. Several ablations are performed to understand the effects of domain randomization, sample complexity, views, and sim-to-real transfer of this method. The data ablation shows that just with a 100 labeled trajectories a 42.01 mAP can be achieved, demonstrating the sample efficiency of using such an approach. Further, behavior cloning models trained on the segmented trajectories outperform a single model trained on the whole trajectory by up to 20%.
ContributorsRaj, Divyanshu (Author) / Gopalan, Nakul (Thesis advisor) / Baral, Chitta (Committee member) / Senanayake, Ransalu (Committee member) / Arizona State University (Publisher)
Created2024
Description
Recently developed large language models have achieved remarkable success on a wide range of natural language tasks. Furthermore, they have been shown to possess an impressive ability to generate fluent and coherent text. Despite all the notable abilities of these models, there exist several efficiency and reliability related challenges. For example, they are vulnerable to a phenomenon called 'hallucination' in which they generate text that is not factually correct and they also have a large number of parameters which makes their inference slow and computationally expensive. With the objective of taking a step closer towards further enabling the widespread adoption of the Natural Language Processing (NLP) systems, this dissertation studies the following question: how to effectively address the efficiency and reliability related concerns of the NLP systems? Specifically, to improve the reliability of models, this dissertation first presents an approach that actively detects and mitigates the hallucinations of LLMs using a retrieval augmented methodology. Note that another strategy to mitigate incorrect predictions is abstention from answering when error is likely, i.e., selective prediction. To this end, I present selective prediction approaches and conduct extensive experiments to demonstrate their effectiveness. Building on top of selective prediction, I also present post-abstention strategies that focus on reliably increasing the coverage of a selective prediction system without considerably impacting its accuracy. Furthermore, this dissertation covers multiple aspects of improving the efficiency including 'inference efficiency' (making model inferences in a computationally efficient manner without sacrificing the prediction accuracy), 'data sample efficiency' (efficiently collecting data instances for training a task-specific system), 'open-domain QA reader efficiency' (leveraging the external knowledge efficiently while answering open-domain questions), and 'evaluation efficiency' (comparing the performance of different models efficiently). In summary, this dissertation highlights several challenges pertinent to the efficiency and reliability involved in the development of NLP systems and provides effective solutions to address them.
ContributorsVarshney, Neeraj (Author) / Baral, Chitta (Thesis advisor) / Yang, Yezhou (Committee member) / Gopalan, Nakul (Committee member) / Banerjee, Pratyay (Committee member) / Arizona State University (Publisher)
Created2024
Description
Teams are increasingly indispensable to achievements in any organizations. Despite the organizations' substantial dependency on teams, fundamental knowledge about the conduct of team-enabled operations is lacking, especially at the {\it social, cognitive} and {\it information} level in relation to team performance and network dynamics. The goal of this dissertation is to create new instruments to {\it predict}, {\it optimize} and {\it explain} teams' performance in the context of composite networks (i.e., social-cognitive-information networks).
Understanding the dynamic mechanisms that drive the success of high-performing teams can provide the key insights into building the best teams and hence lift the productivity and profitability of the organizations. For this purpose, novel predictive models to forecast the long-term performance of teams ({\it point prediction}) as well as the pathway to impact ({\it trajectory prediction}) have been developed. A joint predictive model by exploring the relationship between team level and individual level performances has also been proposed.
For an existing team, it is often desirable to optimize its performance through expanding the team by bringing a new team member with certain expertise, or finding a new candidate to replace an existing under-performing member. I have developed graph kernel based performance optimization algorithms by considering both the structural matching and skill matching to solve the above enhancement scenarios. I have also worked towards real time team optimization by leveraging reinforcement learning techniques.
With the increased complexity of the machine learning models for predicting and optimizing teams, it is critical to acquire a deeper understanding of model behavior. For this purpose, I have investigated {\em explainable prediction} -- to provide explanation behind a performance prediction and {\em explainable optimization} -- to give reasons why the model recommendations are good candidates for certain enhancement scenarios.
Understanding the dynamic mechanisms that drive the success of high-performing teams can provide the key insights into building the best teams and hence lift the productivity and profitability of the organizations. For this purpose, novel predictive models to forecast the long-term performance of teams ({\it point prediction}) as well as the pathway to impact ({\it trajectory prediction}) have been developed. A joint predictive model by exploring the relationship between team level and individual level performances has also been proposed.
For an existing team, it is often desirable to optimize its performance through expanding the team by bringing a new team member with certain expertise, or finding a new candidate to replace an existing under-performing member. I have developed graph kernel based performance optimization algorithms by considering both the structural matching and skill matching to solve the above enhancement scenarios. I have also worked towards real time team optimization by leveraging reinforcement learning techniques.
With the increased complexity of the machine learning models for predicting and optimizing teams, it is critical to acquire a deeper understanding of model behavior. For this purpose, I have investigated {\em explainable prediction} -- to provide explanation behind a performance prediction and {\em explainable optimization} -- to give reasons why the model recommendations are good candidates for certain enhancement scenarios.
ContributorsLi, Liangyue (Author) / Tong, Hanghang (Thesis advisor) / Baral, Chitta (Committee member) / Liu, Huan (Committee member) / Buchler, Norbou (Committee member) / Arizona State University (Publisher)
Created2018
Description
Image Understanding is a long-established discipline in computer vision, which encompasses a body of advanced image processing techniques, that are used to locate (“where”), characterize and recognize (“what”) objects, regions, and their attributes in the image. However, the notion of “understanding” (and the goal of artificial intelligent machines) goes beyond factual recall of the recognized components and includes reasoning and thinking beyond what can be seen (or perceived). Understanding is often evaluated by asking questions of increasing difficulty. Thus, the expected functionalities of an intelligent Image Understanding system can be expressed in terms of the functionalities that are required to answer questions about an image. Answering questions about images require primarily three components: Image Understanding, question (natural language) understanding, and reasoning based on knowledge. Any question, asking beyond what can be directly seen, requires modeling of commonsense (or background/ontological/factual) knowledge and reasoning.
Knowledge and reasoning have seen scarce use in image understanding applications. In this thesis, we demonstrate the utilities of incorporating background knowledge and using explicit reasoning in image understanding applications. We first present a comprehensive survey of the previous work that utilized background knowledge and reasoning in understanding images. This survey outlines the limited use of commonsense knowledge in high-level applications. We then present a set of vision and reasoning-based methods to solve several applications and show that these approaches benefit in terms of accuracy and interpretability from the explicit use of knowledge and reasoning. We propose novel knowledge representations of image, knowledge acquisition methods, and a new implementation of an efficient probabilistic logical reasoning engine that can utilize publicly available commonsense knowledge to solve applications such as visual question answering, image puzzles. Additionally, we identify the need for new datasets that explicitly require external commonsense knowledge to solve. We propose the new task of Image Riddles, which requires a combination of vision, and reasoning based on ontological knowledge; and we collect a sufficiently large dataset to serve as an ideal testbed for vision and reasoning research. Lastly, we propose end-to-end deep architectures that can combine vision, knowledge and reasoning modules together and achieve large performance boosts over state-of-the-art methods.
Knowledge and reasoning have seen scarce use in image understanding applications. In this thesis, we demonstrate the utilities of incorporating background knowledge and using explicit reasoning in image understanding applications. We first present a comprehensive survey of the previous work that utilized background knowledge and reasoning in understanding images. This survey outlines the limited use of commonsense knowledge in high-level applications. We then present a set of vision and reasoning-based methods to solve several applications and show that these approaches benefit in terms of accuracy and interpretability from the explicit use of knowledge and reasoning. We propose novel knowledge representations of image, knowledge acquisition methods, and a new implementation of an efficient probabilistic logical reasoning engine that can utilize publicly available commonsense knowledge to solve applications such as visual question answering, image puzzles. Additionally, we identify the need for new datasets that explicitly require external commonsense knowledge to solve. We propose the new task of Image Riddles, which requires a combination of vision, and reasoning based on ontological knowledge; and we collect a sufficiently large dataset to serve as an ideal testbed for vision and reasoning research. Lastly, we propose end-to-end deep architectures that can combine vision, knowledge and reasoning modules together and achieve large performance boosts over state-of-the-art methods.
ContributorsAditya, Somak (Author) / Baral, Chitta (Thesis advisor) / Yang, Yezhou (Thesis advisor) / Aloimonos, Yiannis (Committee member) / Lee, Joohyung (Committee member) / Li, Baoxin (Committee member) / Arizona State University (Publisher)
Created2018
Description
Reasoning about the activities of cyber threat actors is critical to defend against cyber
attacks. However, this task is difficult for a variety of reasons. In simple terms, it is difficult
to determine who the attacker is, what the desired goals are of the attacker, and how they will
carry out their attacks. These three questions essentially entail understanding the attacker’s
use of deception, the capabilities available, and the intent of launching the attack. These
three issues are highly inter-related. If an adversary can hide their intent, they can better
deceive a defender. If an adversary’s capabilities are not well understood, then determining
what their goals are becomes difficult as the defender is uncertain if they have the necessary
tools to accomplish them. However, the understanding of these aspects are also mutually
supportive. If we have a clear picture of capabilities, intent can better be deciphered. If we
understand intent and capabilities, a defender may be able to see through deception schemes.
In this dissertation, I present three pieces of work to tackle these questions to obtain
a better understanding of cyber threats. First, we introduce a new reasoning framework
to address deception. We evaluate the framework by building a dataset from DEFCON
capture-the-flag exercise to identify the person or group responsible for a cyber attack.
We demonstrate that the framework not only handles cases of deception but also provides
transparent decision making in identifying the threat actor. The second task uses a cognitive
learning model to determine the intent – goals of the threat actor on the target system.
The third task looks at understanding the capabilities of threat actors to target systems by
identifying at-risk systems from hacker discussions on darkweb websites. To achieve this
task we gather discussions from more than 300 darkweb websites relating to malicious
hacking.
attacks. However, this task is difficult for a variety of reasons. In simple terms, it is difficult
to determine who the attacker is, what the desired goals are of the attacker, and how they will
carry out their attacks. These three questions essentially entail understanding the attacker’s
use of deception, the capabilities available, and the intent of launching the attack. These
three issues are highly inter-related. If an adversary can hide their intent, they can better
deceive a defender. If an adversary’s capabilities are not well understood, then determining
what their goals are becomes difficult as the defender is uncertain if they have the necessary
tools to accomplish them. However, the understanding of these aspects are also mutually
supportive. If we have a clear picture of capabilities, intent can better be deciphered. If we
understand intent and capabilities, a defender may be able to see through deception schemes.
In this dissertation, I present three pieces of work to tackle these questions to obtain
a better understanding of cyber threats. First, we introduce a new reasoning framework
to address deception. We evaluate the framework by building a dataset from DEFCON
capture-the-flag exercise to identify the person or group responsible for a cyber attack.
We demonstrate that the framework not only handles cases of deception but also provides
transparent decision making in identifying the threat actor. The second task uses a cognitive
learning model to determine the intent – goals of the threat actor on the target system.
The third task looks at understanding the capabilities of threat actors to target systems by
identifying at-risk systems from hacker discussions on darkweb websites. To achieve this
task we gather discussions from more than 300 darkweb websites relating to malicious
hacking.
ContributorsNunes, Eric (Author) / Shakarian, Paulo (Thesis advisor) / Ahn, Gail-Joon (Committee member) / Baral, Chitta (Committee member) / Cooke, Nancy J. (Committee member) / Arizona State University (Publisher)
Created2018
Description
The goal of fact checking is to determine if a given claim holds. A promising ap- proach for this task is to exploit reference information in the form of knowledge graphs (KGs), a structured and formal representation of knowledge with semantic descriptions of entities and relations. KGs are successfully used in multiple appli- cations, but the information stored in a KG is inevitably incomplete. In order to address the incompleteness problem, this thesis proposes a new method built on top of recent results in logical rule discovery in KGs called RuDik and a probabilistic extension of answer set programs called LPMLN.
This thesis presents the integration of RuDik which discovers logical rules over a given KG and LPMLN to do probabilistic inference to validate a fact. While automatically discovered rules over a KG are for human selection and revision, they can be turned into LPMLN programs with a minor modification. Leveraging the probabilistic inference in LPMLN, it is possible to (i) derive new information which is not explicitly stored in a KG with a probability associated with it, and (ii) provide supporting facts and rules for interpretable explanations for such decisions.
Also, this thesis presents experiments and results to show that this approach can label claims with high precision. The evaluation of the system also sheds light on the role played by the quality of the given rules and the quality of the KG.
This thesis presents the integration of RuDik which discovers logical rules over a given KG and LPMLN to do probabilistic inference to validate a fact. While automatically discovered rules over a KG are for human selection and revision, they can be turned into LPMLN programs with a minor modification. Leveraging the probabilistic inference in LPMLN, it is possible to (i) derive new information which is not explicitly stored in a KG with a probability associated with it, and (ii) provide supporting facts and rules for interpretable explanations for such decisions.
Also, this thesis presents experiments and results to show that this approach can label claims with high precision. The evaluation of the system also sheds light on the role played by the quality of the given rules and the quality of the KG.
ContributorsPradhan, Anish (Author) / Lee, Joohyung (Thesis advisor) / Baral, Chitta (Committee member) / Papotti, Paolo (Committee member) / Arizona State University (Publisher)
Created2018
Description
Knowledge Representation (KR) is one of the prominent approaches to Artificial Intelligence (AI) that is concerned with representing knowledge in a form that computer systems can utilize to solve complex problems. Answer Set Programming (ASP), based on the stable model semantics, is a widely-used KR framework that facilitates elegant and efficient representations for many problem domains that require complex reasoning.
However, while ASP is effective on deterministic problem domains, it is not suitable for applications involving quantitative uncertainty, for example, those that require probabilistic reasoning. Furthermore, it is hard to utilize information that can be statistically induced from data with ASP problem modeling.
This dissertation presents the language LP^MLN, which is a probabilistic extension of the stable model semantics with the concept of weighted rules, inspired by Markov Logic. An LP^MLN program defines a probability distribution over "soft" stable models, which may not satisfy all rules, but the more rules with the bigger weights they satisfy, the bigger their probabilities. LP^MLN takes advantage of both ASP and Markov Logic in a single framework, allowing representation of problems that require both logical and probabilistic reasoning in an intuitive and elaboration tolerant way.
This dissertation establishes formal relations between LP^MLN and several other formalisms, discusses inference and weight learning algorithms under LP^MLN, and presents systems implementing the algorithms. LP^MLN systems can be used to compute other languages translatable into LP^MLN.
The advantage of LP^MLN for probabilistic reasoning is illustrated by a probabilistic extension of the action language BC+, called pBC+, defined as a high-level notation of LP^MLN for describing transition systems. Various probabilistic reasoning about transition systems, especially probabilistic diagnosis, can be modeled in pBC+ and computed using LP^MLN systems. pBC+ is further extended with the notion of utility, through a decision-theoretic extension of LP^MLN, and related with Markov Decision Process (MDP) in terms of policy optimization problems. pBC+ can be used to represent (PO)MDP in a succinct and elaboration tolerant way, which enables planning with (PO)MDP algorithms in action domains whose description requires rich KR constructs, such as recursive definitions and indirect effects of actions.
However, while ASP is effective on deterministic problem domains, it is not suitable for applications involving quantitative uncertainty, for example, those that require probabilistic reasoning. Furthermore, it is hard to utilize information that can be statistically induced from data with ASP problem modeling.
This dissertation presents the language LP^MLN, which is a probabilistic extension of the stable model semantics with the concept of weighted rules, inspired by Markov Logic. An LP^MLN program defines a probability distribution over "soft" stable models, which may not satisfy all rules, but the more rules with the bigger weights they satisfy, the bigger their probabilities. LP^MLN takes advantage of both ASP and Markov Logic in a single framework, allowing representation of problems that require both logical and probabilistic reasoning in an intuitive and elaboration tolerant way.
This dissertation establishes formal relations between LP^MLN and several other formalisms, discusses inference and weight learning algorithms under LP^MLN, and presents systems implementing the algorithms. LP^MLN systems can be used to compute other languages translatable into LP^MLN.
The advantage of LP^MLN for probabilistic reasoning is illustrated by a probabilistic extension of the action language BC+, called pBC+, defined as a high-level notation of LP^MLN for describing transition systems. Various probabilistic reasoning about transition systems, especially probabilistic diagnosis, can be modeled in pBC+ and computed using LP^MLN systems. pBC+ is further extended with the notion of utility, through a decision-theoretic extension of LP^MLN, and related with Markov Decision Process (MDP) in terms of policy optimization problems. pBC+ can be used to represent (PO)MDP in a succinct and elaboration tolerant way, which enables planning with (PO)MDP algorithms in action domains whose description requires rich KR constructs, such as recursive definitions and indirect effects of actions.
ContributorsWang, Yi (Author) / Lee, Joohyung (Thesis advisor) / Baral, Chitta (Committee member) / Kambhampati, Subbarao (Committee member) / Natarajan, Sriraam (Committee member) / Srivastava, Siddharth (Committee member) / Arizona State University (Publisher)
Created2019
Description
For the past three decades, the design of an effective strategy for generating poetry that matches that of a human’s creative capabilities and complexities has been an elusive goal in artificial intelligence (AI) and natural language generation (NLG) research, and among linguistic creativity researchers in particular. This thesis presents a novel approach to fixed verse poetry generation using neural word embeddings. During the course of generation, a two layered poetry classifier is developed. The first layer uses a lexicon based method to classify poems into types based on form and structure, and the second layer uses a supervised classification method to classify poems into subtypes based on content with an accuracy of 92%. The system then uses a two-layer neural network to generate poetry based on word similarities and word movements in a 50-dimensional vector space.
The verses generated by the system are evaluated using rhyme, rhythm, syllable counts and stress patterns. These computational features of language are considered for generating haikus, limericks and iambic pentameter verses. The generated poems are evaluated using a Turing test on both experts and non-experts. The user study finds that only 38% computer generated poems were correctly identified by nonexperts while 65% of the computer generated poems were correctly identified by experts. Although the system does not pass the Turing test, the results from the Turing test suggest an improvement of over 17% when compared to previous methods which use Turing tests to evaluate poetry generators.
The verses generated by the system are evaluated using rhyme, rhythm, syllable counts and stress patterns. These computational features of language are considered for generating haikus, limericks and iambic pentameter verses. The generated poems are evaluated using a Turing test on both experts and non-experts. The user study finds that only 38% computer generated poems were correctly identified by nonexperts while 65% of the computer generated poems were correctly identified by experts. Although the system does not pass the Turing test, the results from the Turing test suggest an improvement of over 17% when compared to previous methods which use Turing tests to evaluate poetry generators.
ContributorsMagge, Arjun (Author) / Syrotiuk, Violet R. (Thesis advisor) / Baral, Chitta (Committee member) / Hogue, Cynthia (Committee member) / Bazzi, Rida (Committee member) / Arizona State University (Publisher)
Created2016
Description
Question Answering has been under active research for decades, but it has recently taken the spotlight following IBM Watson's success in Jeopardy! and digital assistants such as Apple's Siri, Google Now, and Microsoft Cortana through every smart-phone and browser. However, most of the research in Question Answering aims at factual questions rather than deep ones such as ``How'' and ``Why'' questions.
In this dissertation, I suggest a different approach in tackling this problem. We believe that the answers of deep questions need to be formally defined before found.
Because these answers must be defined based on something, it is better to be more structural in natural language text; I define Knowledge Description Graphs (KDGs), a graphical structure containing information about events, entities, and classes. We then propose formulations and algorithms to construct KDGs from a frame-based knowledge base, define the answers of various ``How'' and ``Why'' questions with respect to KDGs, and suggest how to obtain the answers from KDGs using Answer Set Programming. Moreover, I discuss how to derive missing information in constructing KDGs when the knowledge base is under-specified and how to answer many factual question types with respect to the knowledge base.
After having the answers of various questions with respect to a knowledge base, I extend our research to use natural language text in specifying deep questions and knowledge base, generate natural language text from those specification. Toward these goals, I developed NL2KR, a system which helps in translating natural language to formal language. I show NL2KR's use in translating ``How'' and ``Why'' questions, and generating simple natural language sentences from natural language KDG specification. Finally, I discuss applications of the components I developed in Natural Language Understanding.
In this dissertation, I suggest a different approach in tackling this problem. We believe that the answers of deep questions need to be formally defined before found.
Because these answers must be defined based on something, it is better to be more structural in natural language text; I define Knowledge Description Graphs (KDGs), a graphical structure containing information about events, entities, and classes. We then propose formulations and algorithms to construct KDGs from a frame-based knowledge base, define the answers of various ``How'' and ``Why'' questions with respect to KDGs, and suggest how to obtain the answers from KDGs using Answer Set Programming. Moreover, I discuss how to derive missing information in constructing KDGs when the knowledge base is under-specified and how to answer many factual question types with respect to the knowledge base.
After having the answers of various questions with respect to a knowledge base, I extend our research to use natural language text in specifying deep questions and knowledge base, generate natural language text from those specification. Toward these goals, I developed NL2KR, a system which helps in translating natural language to formal language. I show NL2KR's use in translating ``How'' and ``Why'' questions, and generating simple natural language sentences from natural language KDG specification. Finally, I discuss applications of the components I developed in Natural Language Understanding.
ContributorsVo, Nguyen Ha (Author) / Baral, Chitta (Thesis advisor) / Lee, Joohyung (Committee member) / VanLehn, Kurt (Committee member) / Tran, Son Cao (Committee member) / Arizona State University (Publisher)
Created2015