Jonathan Balloch
Atlanta, Georgia, United States
815 followers
500+ connections
Experience
-
-
-
-
-
-
-
-
-
-
-
-
-
Education
Volunteer Experience
-
Mentor, Montgomery Blair High School
FIRST
- 1 year 7 months
Education
Mentor assisting with the first robotics team developing their competition robot and assisting with computer science education of the team as they transition from LabView to Java
Publications
-
Neuro-Symbolic World Models for Adapting to Open World Novelty
AAMAS '23: Proceedings of the 2023 International Conference on Autonomous Agents and Multiagent Systems
Most reinforcement learning (RL) methods assume that the world is a closed, fixed process, when in reality most real world problems are open, changing over time. To address this, we introduce WorldCloner, an end-to-end trainable neuro-symbolic world model that learns an efficient symbolic model of transitions and uses this world model to improve novelty adaptation. We show that the symbolic world model helps WorldCloner adapt its policy more efficiently than neural-only reinforcement learning…
Most reinforcement learning (RL) methods assume that the world is a closed, fixed process, when in reality most real world problems are open, changing over time. To address this, we introduce WorldCloner, an end-to-end trainable neuro-symbolic world model that learns an efficient symbolic model of transitions and uses this world model to improve novelty adaptation. We show that the symbolic world model helps WorldCloner adapt its policy more efficiently than neural-only reinforcement learning methods.
Other authors -
-
NovGrid: A Flexible Grid World for Evaluating Agent Response to Novelty
AAAI 2022 Spring Symposium on Designing Artificial Intelligence for Open Worlds
A robust body of reinforcement learning techniques have been developed to solve complex sequential decision making problems. However, these methods assume that train and evaluation tasks come from similarly or identically distributed environments. This assumption does not hold in real life where small novel changes to the environment can make a previously learned policy fail or introduce simpler solutions that might never be found. To that end we explore the concept of {\em novelty}, defined in…
A robust body of reinforcement learning techniques have been developed to solve complex sequential decision making problems. However, these methods assume that train and evaluation tasks come from similarly or identically distributed environments. This assumption does not hold in real life where small novel changes to the environment can make a previously learned policy fail or introduce simpler solutions that might never be found. To that end we explore the concept of {\em novelty}, defined in this work as the sudden change to the mechanics or properties of environment. We provide an ontology of for novelties most relevant to sequential decision making, which distinguishes between novelties that affect objects versus actions, unary properties versus non-unary relations, and the distribution of solutions to a task. We introduce NovGrid, a novelty generation framework built on MiniGrid, acting as a toolkit for rapidly developing and evaluating novelty-adaptation-enabled reinforcement learning techniques. Along with the core NovGrid we provide exemplar novelties aligned with our ontology and instantiate them as novelty templates that can be applied to many MiniGrid-compliant environments. Finally, we present a set of metrics built into our framework for the evaluation of novelty-adaptation-enabled machine-learning techniques, and show characteristics of a baseline RL model using these metrics.
Other authors -
-
Automated story generation as question-answering
Proc. of the 3rd Workshop on Narrative Understanding
Neural language model-based approaches to automated story generation suffer from two important limitations. First, language model-based story generators generally do not work toward a given goal or ending. Second, they often lose coherence as the story gets longer. We propose a novel approach to automated story generation that treats the problem as one of generative question-answering. Our proposed story generation system starts with sentences encapsulating the final event of the story. The…
Neural language model-based approaches to automated story generation suffer from two important limitations. First, language model-based story generators generally do not work toward a given goal or ending. Second, they often lose coherence as the story gets longer. We propose a novel approach to automated story generation that treats the problem as one of generative question-answering. Our proposed story generation system starts with sentences encapsulating the final event of the story. The system then iteratively (1) analyzes the text describing the most recent event, (2) generates a question about "why" a character is doing the thing they are doing in the event, and then (3) attempts to generate another, preceding event that answers this question.
Other authors -
-
Memory-efficient semi-supervised continual learning: The world is its own replay buffer
2021 International Joint Conference on Neural Networks (IJCNN)
Rehearsal is a critical component for class-incremental continual learning, yet it requires a substantial memory budget. Our work investigates whether we can significantly reduce this memory budget by leveraging unlabeled data from an agent's environment in a realistic and challenging continual learning paradigm. Specifically, we explore and formalize a novel semi-supervised continual learning (SSCL) setting, where labeled data is scarce yet non-i.i.d. unlabeled data from the agent's…
Rehearsal is a critical component for class-incremental continual learning, yet it requires a substantial memory budget. Our work investigates whether we can significantly reduce this memory budget by leveraging unlabeled data from an agent's environment in a realistic and challenging continual learning paradigm. Specifically, we explore and formalize a novel semi-supervised continual learning (SSCL) setting, where labeled data is scarce yet non-i.i.d. unlabeled data from the agent's environment is plentiful. Importantly, data distributions in the SSCL setting are realistic and therefore reflect object class correlations between, and among, the labeled and unlabeled data distributions. We show that a strategy built on pseudo-labeling, consistency regularization, Out-of- Distribution (OoD) detection, and knowledge distillation reduces forgetting in this setting. Our approach, DistillMatch, increases performance over the state-of-the-art by no less than 8.7% average task accuracy and up to 54.5% average task accuracy in SSCL CIFAR-100 experiments. Moreover, we demonstrate that DistillMatch can save up to 0.23 stored images per processed unlabeled image compared to the next best method which only saves 0.08. Our results suggest that focusing on realistic correlated distributions is a significantly new perspective, which accentuates the importance of leveraging the world's structure as a continual learning strategy. Our code is available at https://github.com/GT-RIPL/DistillMatch-SSCL
Other authors -
-
Always be dreaming: A new approach for data-free class-incremental learning
Proceedings of the IEEE/CVF International Conference on Computer Vision
Modern computer vision applications suffer from catastrophic forgetting when incrementally learning new concepts over time. The most successful approaches to alleviate this forgetting require extensive replay of previously seen data, which is problematic when memory constraints or data legality concerns exist. In this work, we consider the high-impact problem of Data-Free Class-Incremental Learning (DFCIL), where an incremental learning agent must learn new concepts over time without storing…
Modern computer vision applications suffer from catastrophic forgetting when incrementally learning new concepts over time. The most successful approaches to alleviate this forgetting require extensive replay of previously seen data, which is problematic when memory constraints or data legality concerns exist. In this work, we consider the high-impact problem of Data-Free Class-Incremental Learning (DFCIL), where an incremental learning agent must learn new concepts over time without storing generators or training data from past tasks. One approach for DFCIL is to replay synthetic images produced by inverting a frozen copy of the learner's classification model, but we show this approach fails for common class-incremental benchmarks when using standard distillation strategies. We diagnose the cause of this failure and propose a novel incremental distillation strategy for DFCIL, contributing a modified cross-entropy training and importance-weighted feature distillation, and show that our method results in up to a 25.1% increase in final task accuracy (absolute difference) compared to SOTA DFCIL methods for common class-incremental benchmarks. Our method even outperforms several standard replay based methods which store a coreset of images.
Other authors -
-
Tool Macgyvering: Tool Construction Using Geometric Reasoning
2019 International Conference on Robotics and Automation (ICRA)
MacGyvering is defined as creating or repairing something in an inventive or improvised way by utilizing objects that are available at hand. In this paper, we explore a subset of Macgyvering problems involving tool construction, i.e., creating tools from parts available in the environment. We formalize the overall problem domain of tool Macgyvering, introducing three levels of complexity for tool construction and substitution problems, and presenting a novel computational framework aimed at…
MacGyvering is defined as creating or repairing something in an inventive or improvised way by utilizing objects that are available at hand. In this paper, we explore a subset of Macgyvering problems involving tool construction, i.e., creating tools from parts available in the environment. We formalize the overall problem domain of tool Macgyvering, introducing three levels of complexity for tool construction and substitution problems, and presenting a novel computational framework aimed at solving one level of the tool Macgyvering problem, specifically contributing a novel algorithm for tool construction based on geometric reasoning. We validate our approach by constructing three tools using a 7-DOF robot arm.
Other authors -
-
Unbiasing semantic segmentation for robot perception using synthetic data feature transfer
Robot perception systems need to perform reliable image segmentation in real-time on noisy, raw perception data. State-of-the-art segmentation approaches use large CNN models and carefully constructed datasets; however, these models focus on accuracy at the cost of real-time inference. Furthermore, the standard semantic segmentation datasets are not large enough for training CNNs without augmentation and are not representative of noisy, uncurated robot perception data. We propose improving the…
Robot perception systems need to perform reliable image segmentation in real-time on noisy, raw perception data. State-of-the-art segmentation approaches use large CNN models and carefully constructed datasets; however, these models focus on accuracy at the cost of real-time inference. Furthermore, the standard semantic segmentation datasets are not large enough for training CNNs without augmentation and are not representative of noisy, uncurated robot perception data. We propose improving the performance of real-time segmentation frameworks on robot perception data by transferring features learned from synthetic segmentation data. We show that pretraining real-time segmentation architectures with synthetic segmentation data instead of ImageNet improves fine-tuning performance by reducing the bias learned in pretraining and closing the \textit{transfer gap} as a result. Our experiments show that our real-time robot perception models pretrained on synthetic data outperform those pretrained on ImageNet for every scale of fine-tuning data examined. Moreover, the degree to which synthetic pretraining outperforms ImageNet pretraining increases as the availability of robot data decreases, making our approach attractive for robotics domains where dataset collection is hard and/or expensive.
-
The Evolution of Titan's detached haze layer near equinox in 2009
Geophysical Research Letters
In modeling radiative transfer in Titan’s upper atmosphere, notably the "detached haze" layer beyond the stratosphere, it was observed that the atmosphere appeared to "collapse" over a course of approximately 4 weeks. The altitude of the detached haze layer and the outer atmosphere was found to have decreased approximately 200 kilometers in this short period of time.
Other authors -
Courses
-
Aerodynamics
MEAM-545
-
Control Theory
ESE-505
-
Electricity and Magnetism
PHYS-252
-
Intro to Optimization
ESE-504
-
Machine Learning
CIS-520
-
Machine Perception
CIS-580
-
Mechatronic Design
MEAM-510
-
Newtonian Mechanics
PHYS-251
-
Quantum Mechanics
PHYS-253
-
Real-Time Embedded Systems
ESE-519
-
Robotics
MEAM-520
-
Statistical Mechanics
PHYS-254
Languages
-
Spanish
-
Other similar profiles
Explore more posts
-
Charlie C. L. Wang
Thrilled to collaborate with Prof. Gang Xu on accelerating the computation of isogeometric analysis (IGA) for topology-consistent models - a rewarding partnership that began during his visit to Delft in 2016. In this recent paper accepted by Journal of Computational Physics, we present IGA-Graph-Net, an innovative isogeometric analysis framework that leverages Graph Neural Networks and introduces a custom loss function to address Dirichlet boundary conditions, achieving significantly improved accuracy over traditional CNN-based methods for isogeometric analysis on complex geometries. By integrating ResNetV2 and PointTransformer, the framework effectively handles B-spline data and complex boundary conditions, demonstrating enhanced performance on partial differential equation datasets. "IGA-Graph-Net: Isogeometric analysis-reuse method based on graph neural networks for topology-consistent models", Journal of Computational Physics, 2024 (https://lnkd.in/erS5q6ZF) "IGA-Reuse-NET: A deep-learning-based isogeometric analysis-reuse approach with topology-consistent parameterization", Computer Aided Geometric Design (https://lnkd.in/e_WiJzFt) "Isogeometric computation reuse method for complex objects with topology-consistent volumetric parameterization", Computer-Aided Design (https://lnkd.in/eZKDsip4) #isogeometricanalysis #machinelearning
1 Comment -
Saurabh Sarkar Ph.D.
This paper introduces a novel method called Self-Discover, which enables large language models (LLMs) to autonomously generate task-specific reasoning structures. Unlike traditional prompting techniques, Self-Discover combines atomic reasoning modules through a meta-reasoning approach, allowing LLMs to tailor their reasoning process to each unique task. To explain the core concept of the "Self-Discover" method simply: Imagine you have a large puzzle to solve. Traditional methods would give you a set of fixed instructions to follow, no matter what the puzzle looks like. The "Self-Discover" method, however, allows the puzzle solver (the LLM) to create its own set of customized instructions based on the specific puzzle at hand. This way, it can more efficiently and accurately solve different types of puzzles by generating and adapting reasoning steps as needed. https://lnkd.in/gmgSe_q4
1 Comment -
Dr. Bilyana Lilly
In his latest post from #TheBatch, Andrew Ng offers his argument on why #California's proposed law SB-1047 is likely to stifle #AI builders and impose unnecessary guardrails on #AI innovation. The law established a “hazardous capability” designation that makes *builders* of AI models liable if someone uses their models to do harm. Andrew argues that laws should not regulate the technology (the tool itself) but its application (how it is implemented). It's a thoughtful read --> DeepLearning.AI #AIgovernance #AIregulation https://lnkd.in/eBD-9rja
-
Edward Y. Chang
Predictions based on Machine Learning and Statistics often rely on maximal likelihood, which can trap LLMs in biases inherited from the majority opinions of training data. This problem has been substantially addressed through conditional statistics, allowing for mitigation of biases and preventing the cancellation of minority perspectives. https://lnkd.in/gMXKqwyB
-
Algorithms MDPI
The #specialissue "Nature-Inspired Algorithms in Machine Learning (2nd Edition)" has published the fifth paper. We thank the Guest Editors Dr. Szymon Łukasik, Dr. Piotr A. Kowalski, and Dr. Rohit Salgotra for their efforts in this special issue! We invite you to read and share the following: https://lnkd.in/g9zMmabs via MDPI #natureinspiredalgorithms #callforpaper
-
es/iode
📃Scientific paper: Global $\mathcal\{L\}^2$ minimization at uniform exponential rate via geometrically adapted gradient descent in Deep Learning Abstract: We consider the scenario of supervised learning in Deep Learning \(DL\) networks, and exploit the arbitrariness of choice in the Riemannian metric relative to which the gradient descent flow can be defined \(a general fact of differential geometry\). In the standard approach to DL, the gradient flow on the space of parameters \(weights and biases\) is defined with respect to the Euclidean metric. Here instead, we choose the gradient flow with respect to the Euclidean metric in the output layer of the DL network. This naturally induces two modified versions of the gradient descent flow in the parameter space, one adapted for the overparametrized setting, and the other for the underparametrized setting. In the overparametrized case, we prove that, provided that a rank condition holds, all orbits of the modified gradient descent drive the $\{\mathcal L\}^2$ cost to its global minimum at a uniform exponential convergence rate; one thereby obtains an a priori stopping time for any prescribed proximity to the global minimum. We point out relations of the latter to sub-Riemannian geometry. Moreover, we generalize the above framework to the situation in which the rank condition does not hold; in particular, we show that local equilibria can only exist if a rank loss occurs, and that generically, they are not isolated points, but elements of a critical submanifold of parameter space. ;Comment: AMS Latex, 20 pages. Significantly edited and extended, abstract changed Continued on ES/IODE ➡️ https://etcse.fr/8W ------- If you find this interesting, feel free to follow, comment and share. We need your help to enhance our visibility, so that our platform continues to serve you.
-
Junho Song
The optimal design of a structural system poses significant challenges, such as the high dimensionality of random variable space and the cost of computational simulations repeated during optimization. In this new paper, co-authored by Dr. Jungho Kim at UC Berkeley (Kim Jungho; a former Ph.D. student of my research group at Seoul National University) and Prof. Ziqi Wang (Ziqi Wang) at UC Berkeley, we proposed to adaptively train a heteroscedastic Gaussian process based surrogate model described in a low dimensional active subspace for accurate and efficient optimization of structures. The proposed method was successfully demonstrated and tested by three examples: (1) nonlinear mathematical functions of up to 100 random variables, (2) a space truss structure described by 57 random variables, and (3) a steel lattice transmission tower subjected to seismic excitations described by 100 random variables. Congratulations, Dr. Kim!
2 Comments -
Colleen Farrelly
Heterogeneous graph data is ubiquitous; different edge relationships and attribute types come up in social/epidemic graphs, molecular/protein graphs, and buyer behavior graphs. This recent paper addresses heterogeny in graph neural network prediction through a topological tool called sheafs, which aggregate properties of edges and vertices according to a pre-defined (or learned) function. Definitely have a look at this recent arXiv paper out of Cambridge if you are working with these types of graph neural networks! https://lnkd.in/eZHHhtuN
1 Comment -
We-Seminar-AI
A Distributional Perspective of Reinforcement Learning 🌟 TonKien Check this out: https://lnkd.in/gqmiccTC In this seminar we discuss Distributional Reinforcement Learning (RL), a cutting edge approach to reinforcement learning that models not just expected returns, but the entire distribution of possible returns. In this seminar we will introduce the foundational principles, key methodologies and the most recent advances in distributional RL. About us: 📚 We share knowledge about AI through detailed seminars, covering various domains such as Reinforcement Learning, LLM, MLLM, Vector Search, Computer Vision, Web Crawling, and more. 🌍Our goal is to foster an educational community eager to learn and apply AI technologies across different sectors. 🌟 We definately can make mistake as it's human-being so we are really appreciated reading your feedbacks, feel free to express your opinion!!
-
Ershad Khandker
" Flatiron Institute senior research scientist Shiwei Zhang and his team have utilized the Hubbard model to computationally re-create key features of the superconductivity in materials called cuprates that have puzzled scientists for decades. Superfast hovering trains, long-distance power transmission without energy loss, and quicker MRI scanners — all these incredible technological innovations could be within reach if we could develop a material that conducts electricity without any resistance, or “superconducts,” at approximately room temperature. In a paper recently published in the journal Science, researchers report a breakthrough in our understanding of the origins of superconductivity at relatively high (though still frigid) temperatures. The findings concern a class of superconductors that has puzzled scientists since 1986, called ‘cuprates.’ There was tremendous excitement when cuprate superconductors were discovered [in 1986], but no understanding of why they remain superconductive at such high temperatures,” says Shiwei Zhang, a senior research scientist at the Flatiron Institute’s Center for Computational Quantum Physics (CCQ). “I think it’s surprising to everybody that almost 40 years later, we still don’t quite understand why they do what they do. ❗️In the new paper, Zhang and his colleagues successfully re-created features of cuprate superconductivity with a simple model called the two-dimensional Hubbard model, which treats the materials as if they were electrons moving around a quantum chessboard. The breakthrough comes only a few years after the same researchers demonstrated that the simplest version of this model couldn’t perform such a feat. Such straightforward models can spark a deeper understanding of physics, says study co-author Ulrich Schollwöck, a professor at the University of Munich. ❗️👈 The idea in physics is to keep the model as simple as possible because it’s difficult enough on its own,” Schollwöck says. “So in the beginning we studied the simplest version imaginable.” Enhancements to the Hubbard Model In the new study, the researchers added to the 2D Hubbard model the ability for electrons to make diagonal hops, like bishops in chess. With this tweak and thousands of weeks-long simulations on supercomputers, the researchers’ model captured the superconductivity and several other key features of cuprates previously found in experiments. By showing that the humble Hubbard model can describe cuprate superconductivity, the authors prove its worth as a platform for understanding why and how superconductivity emerges." After over 30 years of intense effort by the community without many reliable answers, it’s often been argued that solving the Hubbard model would have to wait for a quantum computer,” Zhang says. “This effort will not only advance research in high-temperature superconductivity, but hopefully also spur more research using ‘classical’ computation to explore the wonders of the quantum world.”
-
Masato Kobayashi
After collecting data approximately five times using ALPHA-α and training with "Bi-ACT," the system can currently operate autonomously as shown (though further verification is needed). Data collection is relatively simple, as each session only takes a few tens of seconds. For reference, the ALOHA paper mentions requiring about 50 iterations. During autonomous operation, input updates are optimized to run at about 100 Hz, while current control operates at 1000 Hz. "ALPHA-α and Bi-ACT Are All You Need: Importance of Position and Force Information/Control for Imitation Learning of Unimanual and Bimanual Robotic Manipulation with Low-Cost System" 🤖 project page: https://lnkd.in/g7cTpuHB 💡 X(more details): https://lnkd.in/g6cqtDu7
-
Enrico Camporeale
𝗚𝗣𝗨 𝘀𝗵𝗮𝗿𝗲 𝗽𝗿𝗼𝗴𝗿𝗮𝗺 at the Deep Learning Lab, SWx-TREC (University of Colorado) We solicit applications to get access to GPU compute time for training 𝗠𝗮𝗰𝗵𝗶𝗻𝗲 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴 𝗺𝗼𝗱𝗲𝗹𝘀 𝗳𝗼𝗿 𝗦𝗽𝗮𝗰𝗲 𝗪𝗲𝗮𝘁𝗵𝗲𝗿/𝗦𝗽𝗮𝗰𝗲 𝗣𝗵𝘆𝘀𝗶𝗰𝘀 𝗮𝗽𝗽𝗹𝗶𝗰𝗮𝘁𝗶𝗼𝗻𝘀. Please send us your research ideas in a two-pager document that includes: • A brief description of the project, including details of the ML method to be trained, its appropriateness, expected deliverables and metrics for success; • Resources required (how many NVIDIA A100 GPUs, for how long, how much storage space; how many CPUs?) We expect a typical (initial) allocation to be 2 GPUs for ~ 3 months (~ 2200 GPU/hours). Storage will be considered on case-by-case (considering that the most popular ML-Helio databases are already stored on our server) • Number of accounts required (how many people need access to the server); • Proficiency and previous experience of team members (will you require additional help to set up the codes, running experiments, etc?). 𝗪𝗵𝗮𝘁 𝗱𝗼 𝘆𝗼𝘂 𝗴𝗲𝘁? Unlimited access to the resources agreed on (i.e., 2x GPUs, 50 Tb of storage) for the duration agreed on (i.e., two months). If your project requires a combination of GPU and CPUs please let us know. 𝗪𝗵𝗮𝘁 𝗱𝗼𝗲𝘀 𝗦𝗪𝘅-𝗧𝗥𝗘𝗖 𝗴𝗲𝘁? We require that all datasets used and produced during the project and all associated codes (pre-processing, ML training, visualization, etc.) will be shared with us through a non-exclusive license We require that at least one representative of our group is invited to regular project meetings (i.e. biweekly update meetings) and that the project will be carried out in a collaborative rather than competitive spirit. We request that the AGU Ethical Obligations and Authorship Policy will be followed, and that SWx TREC/DLL is acknowledged in any papers, posters, or talks given based on the products of the research. https://lnkd.in/gdUw2tiP
1 Comment -
Amarda Shehu
There is often an unspoken assumption by ML researchers that all we need to make progress and model every aspect of our world is data. We do not ascribe to this in my lab. In fact, we believe that data will never be enough. Our experiences with the nuances and complexities of scientific problems have informed us to the insufficiency of data to capture continuous physical processes, which afterall govern our biological and physical world. An example of this is this series of two papers led by my wonderful PhD student, Anowarul Kabir and advanced by a precious multi-year collaboration of my lab with Los Alamos National Lab: Anowarul Kabir, Manish Bhattarai, Kim Rasmussen, Amarda Shehu, Anny Usheva, Alan R Bishop, and Boian S Alexandrov. Examining DNA Breathing with pyDNA-EPBD. Bioinformatics 39(11):btad699, 2023. https://lnkd.in/gZaHE4hS Anowarul Kabir, Manish Bhattarai, Selma Peterson, Yonatan Najman-Licht, Kim Ø Rasmussen, Amarda Shehu, Alan R Bishop, Boian Alexandrov, Anny Usheva. DNA Breathing Integration with Deep learning Foundational Model Advances Genome-wide Binding Prediction of Human Transcription Factors. Nucleic Acids Research: gkae783, 2024. https://lnkd.in/gfTTqcPy Our goal: advance an exceptionally challenging problem in molecular biology, prediction of transcription factor binding sites. Our first step: capture the underlying physics that is missing in the data. Our second step: integrate that now with the data we have in a foundation model for predicting transcription factor binding sites. Performance improves. Most importantly, when we look at the sequence motifs that constitute "signatures" of what makes a transcription factor binding site, we obtain answers. All in the open, nothing opaque.
1 Comment -
Rabab Alomairy, Ph.D.
Are you facing hurdles in scaling up your computational tasks or fine-tuning deep learning models on distributed memory systems with GPU acceleration? Join us at JuliaCon 2024 for the Dagger.jl Workshop! Discover how Dagger's parallel task-based runtime can empower you to efficiently parallelize across threads, nodes, and GPUs. Secure your spot now! 🌟 JuliaCon 2024 #JuliaLang #JuliaCon2024 #ParallelComputing #Daggerjl #GPUComputing
Explore collaborative articles
We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.
Explore MoreOthers named Jonathan Balloch
1 other named Jonathan Balloch is on LinkedIn
See others named Jonathan Balloch