Search

Handwritten font generation is important for preserving cultural heritage and creating personalized designs. It adds an authentic and expressive touch to printed materials, making them visually appealing and establishing a stronger connection with the audience. This paper aims to design a framework for generating handwritten fonts in the Gujarati script, mimicking the variation of human handwriting. The proposed font generation model consists of a learning phase and a generation phase. In the learning phase, Gujarati scripts are analyzed, and rules for designing each character are formulated. This ruleset involves the concatenation of strokes in a stroke-based manner, ensuring visual consistency in the resulting glyphs. The generation phase involves the user providing a small subset of characters, and the system automatically generates the remaining character glyphs based on extracted strokes and learned rules, resulting in handwritten Gujarati fonts. The resulting character glyphs are converted into an open-type font using the FontForge tool, making them compatible with any Gujarati editor. Both subjective and objective evaluations are conducted to assess the synthesized images and fonts. Subjective evaluation through user studies provides feedback on quality and visual appeal, achieving an overall accuracy of 84.84%. Notably, eleven characters demonstrated a success ratio above 90%. Objective evaluation using an existing recognition system achieves an overall accuracy of 84.28% in OCR evaluation. Notably, fifteen characters had a success ratio of 80% or higher., Comment: 13 pages, 2 column, 12 figures

Hallucination is a well-known phenomenon in text generated by large language models (LLMs). The existence of hallucinatory responses is found in almost all application scenarios e.g., summarization, question-answering (QA) etc. For applications requiring high reliability (e.g., customer-facing assistants), the potential existence of hallucination in LLM-generated text is a critical problem. The amount of hallucination can be reduced by leveraging information retrieval to provide relevant background information to the LLM. However, LLMs can still generate hallucinatory content for various reasons (e.g., prioritizing its parametric knowledge over the context, failure to capture the relevant information from the context, etc.). Detecting hallucinations through automated methods is thus paramount. To facilitate research in this direction, we introduce a sophisticated dataset, DelucionQA, that captures hallucinations made by retrieval-augmented LLMs for a domain-specific QA task. Furthermore, we propose a set of hallucination detection methods to serve as baselines for future works from the research community. Analysis and case study are also provided to share valuable insights on hallucination phenomena in the target scenario., Comment: Accepted in EMNLP 2023 (Findings)

Recent approaches have explored language-guided classifiers capable of classifying examples from novel tasks when provided with task-specific natural language explanations, instructions or prompts (Sanh et al., 2022; R. Menon et al., 2022). While these classifiers can generalize in zero-shot settings, their task performance often varies substantially between different language explanations in unpredictable ways (Lu et al., 2022; Gonen et al., 2022). Also, current approaches fail to leverage unlabeled examples that may be available in many scenarios. Here, we introduce TALC, a framework that uses data programming to adapt a language-guided classifier for a new task during inference when provided with explanations from multiple teachers and unlabeled test examples. Our results show that TALC consistently outperforms a competitive baseline from prior work by an impressive 9.3% (relative improvement). Further, we demonstrate the robustness of TALC to variations in the quality and quantity of provided explanations, highlighting its potential in scenarios where learning from multiple teachers or a crowd is involved. Our code is available at: https://github.com/WeiKangda/TALC.git.

Generalized quantifiers (e.g., few, most) are used to indicate the proportions predicates are satisfied (for example, some apples are red). One way to interpret quantifier semantics is to explicitly bind these satisfactions with percentage scopes (e.g., 30%-40% of apples are red). This approach can be helpful for tasks like logic formalization and surface-form quantitative reasoning (Gordon and Schubert, 2010; Roy et al., 2015). However, it remains unclear if recent foundation models possess this ability, as they lack direct training signals. To explore this, we introduce QuRe, a crowd-sourced dataset of human-annotated generalized quantifiers in Wikipedia sentences featuring percentage-equipped predicates. We explore quantifier comprehension in language models using PRESQUE, a framework that combines natural language inference and the Rational Speech Acts framework. Experimental results on the HVD dataset and QuRe illustrate that PRESQUE, employing pragmatic reasoning, performs 20% better than a literal reasoning baseline when predicting quantifier percentage scopes, with no additional training required., Comment: EMNLP 2023

Understanding the internal reasoning behind the predictions of machine learning systems is increasingly vital, given their rising adoption and acceptance. While previous approaches, such as LIME, generate algorithmic explanations by attributing importance to input features for individual examples, recent research indicates that practitioners prefer examining language explanations that explain sub-groups of examples. In this paper, we introduce MaNtLE, a model-agnostic natural language explainer that analyzes multiple classifier predictions and generates faithful natural language explanations of classifier rationale for structured classification tasks. MaNtLE uses multi-task training on thousands of synthetic classification tasks to generate faithful explanations. Simulated user studies indicate that, on average, MaNtLE-generated explanations are at least 11% more faithful compared to LIME and Anchors explanations across three tasks. Human evaluations demonstrate that users can better predict model behavior using explanations from MaNtLE compared to other techniques, Comment: 17 pages, 13 figures, 6 tables

Understanding the spatio-temporal evolution of thermal gradient (G) at and velocity (R) of the solid-liquid and liquid-vapor interfaces is critical for the control of site-specific microstructures in additive manufacturing. In-situ dynamic x-ray radiography (DXR) has been used in recent years to probe the evolution of R with high spatial and temporal resolutions. However, the current methods used to measure the temperature (and therefore, G) are inadequate for sub-melt-pool surface measurement (e.g. thermography) or have lower resolution or limited field-of-view (e.g. x-ray diffraction). In this study, we demonstrate a novel approach to estimate the sub-surface temperature distribution and its time evolution with significantly higher resolutions using DXR data. This methodology uses the Beer-Lambert's law as a physical basis and is demonstrated using an in-situ laser spot-melting experiment on Ti-6Al-4V alloy.

A hallmark of human intelligence is the ability to learn new concepts purely from language. Several recent approaches have explored training machine learning models via natural language supervision. However, these approaches fall short in leveraging linguistic quantifiers (such as 'always' or 'rarely') and mimicking humans in compositionally learning complex tasks. Here, we present LaSQuE, a method that can learn zero-shot classifiers from language explanations by using three new strategies - (1) modeling the semantics of linguistic quantifiers in explanations (including exploiting ordinal strength relationships, such as 'always' > 'likely'), (2) aggregating information from multiple explanations using an attention-based mechanism, and (3) model training via curriculum learning. With these strategies, LaSQuE outperforms prior work, showing an absolute gain of up to 7% in generalizing to unseen real-world classification tasks., Comment: Work in progress

Supervised learning has traditionally focused on inductive learning by observing labeled examples of a task. In contrast, humans have the ability to learn new concepts from language. Here, we explore training zero-shot classifiers for structured data purely from language. For this, we introduce CLUES, a benchmark for Classifier Learning Using natural language ExplanationS, consisting of a range of classification tasks over structured data along with natural language supervision in the form of explanations. CLUES consists of 36 real-world and 144 synthetic classification tasks. It contains crowdsourced explanations describing real-world tasks from multiple teachers and programmatically generated explanations for the synthetic tasks. To model the influence of explanations in classifying an example, we develop ExEnt, an entailment-based model that learns classifiers using explanations. ExEnt generalizes up to 18% better (relative) on novel tasks than a baseline that does not use explanations. We delineate key challenges for automated learning from explanations, addressing which can lead to progress on CLUES in the future. Code and datasets are available at: https://clues-benchmark.github.io., Comment: ACL 2022 (25 pages, 16 figures)

For multiple emergencies caused by natural disasters, it is crucial to allocate resources equitably to each emergency location, especially when the availability of resources is limited in quantity. This paper has developed a multi-event crisis management system using a non-cooperative, complete information, strategic form game model. In the proposed system, each emergency event is assumed to occur in different locations simultaneously. These locations are represented as the players in the game, competing with the other players for an optimal allocation of scarce resources available at different resource stations. The players incur a non-monetary cost for obtaining resource units. The objective of the proposed game is to derive optimal strategies for an effective and fair allocation of resources to the respective players., Comment: Presented in 4th World Congress on Disaster Management, IIT Bombay, Mumbai, India, 29 Jan-01 Feb 2019 (11 pages)

Attacks on a set of agents with cooperative and antagonistic interactions attempting to achieve linear bipartite consensus are considered here. In bipartite consensus, two clusters are formed, agents in each cluster converging to a final state which is negative of the other cluster's final state. The adversary seeks to slow down the bipartite consensus by a Denial of Service (DoS) type attack where the attacker has the capability to break a specific number of links at each time instant. The problem is formulated as an optimal control problem and the optimal strategy for the adversary is determined., Comment: 5 pages, 5 figures

Recently, pre-trained language models (LMs) have achieved strong performance when fine-tuned on difficult benchmarks like SuperGLUE. However, performance can suffer when there are very few labeled examples available for fine-tuning. Pattern Exploiting Training (PET) is a recent approach that leverages patterns for few-shot learning. However, PET uses task-specific unlabeled data. In this paper, we focus on few-shot learning without any unlabeled data and introduce ADAPET, which modifies PET's objective to provide denser supervision during fine-tuning. As a result, ADAPET outperforms PET on SuperGLUE without any task-specific unlabeled data. Our code can be found at https://github.com/rrmenon10/ADAPET., Comment: EMNLP 2021 (12 pages, 2 figures)

The simulation of vehicular wireless channels using geometry-based radio channel models is computationally intensive when the number of scatterers is significantly high. In this paper, we propose a new geometry-based stochastic channel model to simulate and analyze the aforementioned channels based on a framework developed from the theory of ambit processes. Under reasonable assumptions, the underlying mathematical structure of the proposed channel model enables the characterization of high mobility channels in terms of fading statistics, spatio-temporal channel correlation, and Doppler spectrum, besides ensuring tractable analysis. The developed algorithm facilitates fast simulation of high mobility channels and accounts for key features of vehicular channels including appearance and disappearance of multi-path components, spatial consistency, and captures the correlation between time-evolving delay and Doppler associated with multi-path components. Finally, we carry out simulations to obtain crucial insights about the characteristics of typical vehicle-to-infrastructure channels based on the proposed channel model.

Millimeter wave (mmWave) communications which essentially employ directional antennas find applications spanning from indoor short range wireless personal area networks to outdoor cellular networks. A thorough understanding of mmWave signal propagation through the wireless channel provides valuable insights for the design of such networks which in turn dictates the achievable performance limits. High path loss, penetration loss, and negligible signal scattering are certain distinctive features of the mmWave channel which necessitates the development of new mathematical models. Importantly, the impact of directional antennas which spatially filter multi-path components needs to be embodied as an integral part of the channel model. In this paper, we model outdoor directional non-line-of-sight mmWave channels using a combination of stochastic geometry and image theory by expressing channel parameters as joint functions of transmitter-receiver separation distance, antenna half power beamwidth, and antenna beam pointing direction. By approximating the signal propagation path due to first and second order reflections from buildings, closed form analytical expressions for average number of first order reflection components, path loss, and root-mean square delay spread of channel are derived. The accuracy of the model is verified by comparing numerically obtained results with experimental data reported by various urban outdoor peer-to-peer mmWave measurements, thus demonstrating the usefulness of the proposed analytical model for performance evaluation of mmWave communication networks., Comment: 14 pages

Research in Artificial Intelligence is breaking technology barriers every day. New algorithms and high performance computing are making things possible which we could only have imagined earlier. Though the enhancements in AI are making life easier for human beings day by day, there is constant fear that AI based systems will pose a threat to humanity. People in AI community have diverse set of opinions regarding the pros and cons of AI mimicking human behavior. Instead of worrying about AI advancements, we propose a novel idea of cognitive agents, including both human and machines, living together in a complex adaptive ecosystem, collaborating on human computation for producing essential social goods while promoting sustenance, survival and evolution of the agents' life cycle. We highlight several research challenges and technology barriers in achieving this goal. We propose a governance mechanism around this ecosystem to ensure ethical behaviors of all cognitive agents. Along with a novel set of use-cases of Cogniculture, we discuss the road map ahead for this journey.

Deep Reinforcement Learning has been able to achieve amazing successes in a variety of domains from video games to continuous control by trying to maximize the cumulative reward. However, most of these successes rely on algorithms that require a large amount of data to train in order to obtain results on par with human-level performance. This is not feasible if we are to deploy these systems on real world tasks and hence there has been an increased thrust in exploring data efficient algorithms. To this end, we propose the Shared Learning framework aimed at making $Q$-ensemble algorithms data-efficient. For achieving this, we look into some principles of transfer learning which aim to study the benefits of information exchange across tasks in reinforcement learning and adapt transfer to learning our value function estimates in a novel manner. In this paper, we consider the special case of transfer between the value function estimates in the $Q$-ensemble architecture of BootstrappedDQN. We further empirically demonstrate how our proposed framework can help in speeding up the learning process in $Q$-ensembles with minimum computational overhead on a suite of Atari 2600 Games., Comment: Submitted to AAAI 2018

The analytical model for resource utilization and the variable node time and conditional node model for the enhanced ATAMM model for a real-time data flow architecture are presented in this research. The Algorithm To Architecture Mapping Model, ATAMM, is a Petri net based graph theoretic model developed at Old Dominion University, and is capable of modeling the execution of large-grained algorithms on a real-time data flow architecture. Using the resource utilization model, the resource envelope may be obtained directly from a given graph and, consequently, the maximum number of required resources may be evaluated. The node timing diagram for one iteration period may be obtained using the analytical resource envelope. The variable node time model, which describes the change in resource requirement for the execution of an algorithm under node time variation, is useful to expand the applicability of the ATAMM model to heterogeneous architectures. The model also describes a method of detecting the presence of resource limited mode and its subsequent prevention. Graphs with conditional nodes are shown to be reduced to equivalent graphs with time varying nodes and, subsequently, may be analyzed using the variable node time model to determine resource requirements. Case studies are performed on three graphs for the illustration of applicability of the analytical theories.