Tim Oates

NeurIPS Conference 2024 Conference Paper

A Walsh Hadamard Derived Linear Vector Symbolic Architecture

• Mohammad Mahmudul Alam
• Alexander Oberle
• Edward Raff
• Stella Biderman
• Tim Oates
• James Holt

Vector Symbolic Architectures (VSAs) are one approach to developing Neuro-symbolic AI, where two vectors in $\mathbb{R}^d$ are 'bound' together to produce a new vector in the same space. VSAs support the commutativity and associativity of this binding operation, along with an inverse operation, allowing one to construct symbolic-style manipulations over real-valued vectors. Most VSAs were developed before deep learning and automatic differentiation became popular and instead focused on efficacy in hand-designed systems. In this work, we introduce the Hadamard-derived linear Binding (HLB), which is designed to have favorable computational efficiency, and efficacy in classic VSA tasks, and perform well in differentiable systems.

AAAI Conference 2023 Short Paper

RFC-Net: Learning High Resolution Global Features for Medical Image Segmentation on a Computational Budget (Student Abstract)

• Sourajit Saha
• Shaswati Saha
• Md Osman Gani
• Tim Oates
• David Chapman

Learning High-Resolution representations is essential for semantic segmentation. Convolutional neural network (CNN) architectures with downstream and upstream propagation flow are popular for segmentation in medical diagnosis. However, due to performing spatial downsampling and upsampling in multiple stages, information loss is inexorable. On the contrary, connecting layers densely on high spatial resolution is computationally expensive. In this work, we devise a Loose Dense Connection Strategy to connect neurons in subsequent layers with reduced parameters. On top of that, using a m-way Tree structure for feature propagation we propose Receptive Field Chain Network (RFC-Net) that learns high-resolution global features on a compressed computational space. Our experiments demonstrates that RFC Net achieves state-of-the-art performance on Kvasir and CVC-ClinicDB benchmarks for Polyp segmentation. Our code is publicly available at github.com/sourajitcs/RFC-NetAAAI23.

AAAI Conference 2021 Conference Paper

Bringing UMAP Closer to the Speed of Light with GPU Acceleration

• Corey J. Nolet
• Victor Lafargue
• Edward Raff
• Thejaswi Nanditale
• Tim Oates
• John Zedlewski
• Joshua Patterson

The Uniform Manifold Approximation and Projection (UMAP) algorithm has become widely popular for its ease of use, quality of results, and support for exploratory, unsupervised, supervised, and semi-supervised learning. While many algorithms can be ported to a GPU in a simple and direct fashion, such efforts have resulted in inefficient and inaccurate versions of UMAP. We show a number of techniques that can be used to make a faster and more faithful GPU version of UMAP, and obtain speedups of up to 100x in practice. Many of these design choices/lessons are general purpose and may inform the conversion of other graph and manifold learning algorithms to use GPUs. Our implementation has been made publicly available as part of the open source RAPIDS cuML library (https: //github. com/rapidsai/cuml).

NeurIPS Conference 2021 Conference Paper

Learning with Holographic Reduced Representations

• Ashwinkumar Ganesan
• Hang Gao
• Sunil Gandhi
• Edward Raff
• Tim Oates
• James Holt
• Mark McLean

Holographic Reduced Representations (HRR) are a method for performing symbolic AI on top of real-valued vectors by associating each vector with an abstract concept, and providing mathematical operations to manipulate vectors as if they were classic symbolic objects. This method has seen little use outside of older symbolic AI work and cognitive science. Our goal is to revisit this approach to understand if it is viable for enabling a hybrid neural-symbolic approach to learning as a differential component of a deep learning architecture. HRRs today are not effective in a differential solution due to numerical instability, a problem we solve by introducing a projection step that forces the vectors to exist in a well behaved point in space. In doing so we improve the concept retrieval efficacy of HRRs by over $100\times$. Using multi-label classification we demonstrate how to leverage the symbolic HRR properties to develop a output layer and loss function that is able to learn effectively, and allows us to investigate some of the pros and cons of an HRR neuro-symbolic learning approach.

AAMAS Conference 2019 Conference Paper

Learning Behaviors from a Single Video Demonstration Using Human Feedback

• Sunil Gandhi
• Tim Oates
• Tinoosh Mohsenin
• Nicholas R. Waytowich

In this paper we present a method for learning from video demonstrations by using human feedback to construct a mapping between the internal state representation of the agent and the visual representation from the video. In this way, we leverage the advantages of both these representations, i. e. , we learn the policy using agent centered state representations, but are able to specify the expected behavior using video demonstrations. We show the effectiveness of our method by teaching a hopper agent in the MuJoCo simulator to perform a backflip using a single video demonstration generated in MuJoCo as well as from a real-world YouTube video of a person performing a backflip.

TAAS Journal 2017 Journal Article

Feature Construction for Controlling Swarms by Visual Demonstration

• Karan K. Budhraja
• John Winder
• Tim Oates

Agent-based modeling is a paradigm of modeling dynamic systems of interacting agents that are individually governed by specified behavioral rules. Training a model of such agents to produce an emergent behavior by specification of the emergent (as opposed to agent) behavior is easier from a demonstration perspective. While many approaches involve manual behavior specification via code or reliance on a defined taxonomy of possible behaviors, the meta-modeling framework in Miner [2010] generates mapping functions between agent-level parameters and swarm-level parameters, which are re-usable once generated. This work builds on that framework by integrating demonstration by image or video. The demonstrator specifies spatial motion of the agents over time and retrieves agent-level parameters required to execute that motion. The framework, at its core, uses computationally cheap image-processing algorithms. Our work is tested with a combination of primitive visual feature extraction methods (contour area and shape) and features generated using a pre-trained deep neural network in different stages of image featurization. The framework is also evaluated for its potential using complex visual features for all image featurization stages. Experimental results show significant coherence between demonstrated behavior and predicted behavior based on estimated agent-level parameters specific to the spatial arrangement of agents.

AAAI Conference 2016 Conference Paper

Adaptive Normalized Risk-Averting Training for Deep Neural Networks

• Zhiguang Wang
• Tim Oates
• James Lo

This paper proposes a set of new error criteria and a learning approach, called Adaptive Normalized Risk- Averting Training (ANRAT) to attack the non-convex optimization problem in training deep neural networks without pretraining. Theoretically, we demonstrate its effectiveness based on the expansion of the convexity region. By analyzing the gradient on the convexity index λ, we explain the reason why our learning method using gradient descent works. In practice, we show how this training method is successfully applied for improved training of deep neural networks to solve visual recognition tasks on the MNIST and CIFAR- 10 datasets. Using simple experimental settings without pretraining and other tricks, we obtain results comparable or superior to those reported in recent literature on the same tasks using standard ConvNets + MSE/cross entropy. Performance on deep/shallow multilayer perceptron and Denoised Auto-encoder is also explored. ANRAT can be combined with other quasi- Newton training methods, innovative network variants, regularization techniques and other common tricks in DNNs. Other than unsupervised pretraining, it provides a new perspective to address the non-convex optimization strategy in training DNNs.

IJCAI Conference 2015 Conference Paper

Imaging Time-Series to Improve Classification and Imputation

• Zhiguang Wang
• Tim Oates

Inspired by recent successes of deep learning in computer vision, we propose a novel framework for encoding time series as different types of images, namely, Gramian Angular Summation/Difference Fields (GASF/GADF) and Markov Transition Fields (MTF). This enables the use of techniques from computer vision for time series classification and imputation. We used Tiled Convolutional Neural Networks (tiled CNNs) on 20 standard datasets to learn high-level features from the individual and compound GASF-GADF-MTF images. Our approaches achieve highly competitive results when compared to nine of the current best time series classification approaches. Inspired by the bijection property of GASF on 0/1 rescaled data, we train Denoised Auto-encoders (DA) on the GASF images of four standard and one synthesized compound dataset. The imputation MSE on test data is reduced by 12. 18%-48. 02% when compared to using the raw data. An analysis of the features and weights learned via tiled CNNs and DAs explains why the approaches work.

IJCAI Conference 2009 Conference Paper

• Fusun Yaman
• Tim Oates
• Mark Burstein

Existing work on workflow mining ignores the dataflow aspect of the problem. This is not acceptable for service-oriented applications that use Web services with typed inputs and outputs. We propose a novel algorithm WIT (Workflow Inference from Traces) which identifies the context similarities of the observed actions based on the dataflow and uses model merging techniques to generalize the control flow and the dataflow simultaneously. We identify the class of workflows that WIT can learn correctly. We implemented WIT and tested it on a real world medical scheduling domain where WIT was able to find a good approximation of the target workflow.

IJCAI Conference 2007 Conference Paper

• Aarti Gupta
• Tim Oates

A variety of text processing tasks require or benefit from semantic resources such as ontologies and lexicons. Creating these resources manually is tedious, time consuming, and prone to error. We present a new algorithm for using the web to determine the correct concept in an existing ontology to lexicalize previously unknown words, such as might be discovered while processing texts. A detailed empirical comparison of our algorithm with two existing algorithms (Cilibrasi & Vitanyi 2004, Maedche et al. 2002) is described, leading to insights into the sources of the algorithms' strengths and weaknesses.

AIJ Journal 2006 Journal Article

Book review

• Tim Oates
• Waiyian Chong

AIJ Journal 2005 Journal Article

Book review

• Tim Oates

IJCAI Conference 2005 Conference Paper

Transfer in Learning by Doing

• Bill Krueger
• Tim Oates
• Tom Armstrong
• Paul Cohen
• Carole

We develop two related themes, learning procedures and knowledge transfer. This paper introduces two methods for learning procedures and one for transferring previously-learned knowledge to a slightly different task. We demonstrate by experiment that transfer accelerates learning.

AAAI Conference 2004 Conference Paper

On the Relationship between Lexical Semantics and Syntax for the Inference of Context-Free Grammars

• Tim Oates
• Justin Harris

Context-free grammars cannot be identified in the limit from positive examples (Gold 1967), yet natural language grammars are more powerful than context-free grammars and humans learn them with remarkable ease from positive examples (Marcus 1993). Identifiability results for formal languages ignore a potentially powerful source of information available to learners of natural languages, namely, meanings. This paper explores the learnability of syntax (i. e. contextfree grammars) given positive examples and knowledge of lexical semantics, and the learnability of lexical semantics given knowledge of syntax. The long-term goal is to develop an approach to learning both syntax and semantics that bootstraps itself, using limited knowledge about syntax to infer additional knowledge about semantics, and limited knowledge about semantics to infer additional knowledge about syntax.

AAAI Conference 2000 Conference Paper

A Method for Clustering the Experiences of a Mobile Robot that Accords with Human Judgments

• Tim Oates
• and Paul R. Cohen

If robotic agents are to act autonomously they must have the ability to construct and reason about models of their physical environment. For example, planning to achieve goals requires knowledge of how the robot’s actions affect the state of the world over time. The traditional approach of handcoding this knowledge is often quite difficult, especially for robotic agents with rich sensing abilities that exist in dynamic and uncertain environments. Ideally, robots would acquire knowledge of their environment and then use this knowledge to act. We present an unsupervised learning method that allows a robotic agent to identify and represent qualitatively different outcomes of actions. Experiments with a Pioneer-1 mobile robot demonstrate the utility of the approach with respect to capturing the structure and dynamics of a complex, real-world environment, and show that the models acquired by the robot correlate surprisingly well with human models of the environment.

IJCAI Conference 1999 Conference Paper

Efficient Mining of Statistical Dependencies

• Tim Oates
• Matthew D. Schmill
• Paul R Cohen

The Multi-Stream Dependency Detection algorithm finds rules that capture statistical dependencies between patterns in multivariate time series of categorical data [Oates and Cohen, 1996c]. Rule strength is measured by the G statistic [Wickens, 1989], and an upper bound on the value of G for the descendants of a node allows MSDD'S search space to be pruned. However, in the worst case, the algorithm will explore exponentially many rules. This paper presents and empirically evaluates two ways of addressing this problem. The first is a set of three methods for reducing the size of MSDD'S search space based on information collected during the search process. Second, we discuss an implementation of MSDD that distributes its computations over multiple machines on a network.

AAAI Conference 1999 Conference Paper

Toward a Theoretical Understanding of Why and When Decision Tree Pruning Algorithms Fail

• Tim Oates
• David Jensen
• University of Massachusetts

Recent empirical studies revealed two surprising pathologies of several common decision tree pruning algorithms. First, tree size is often a linear function of training set size, evenwhenadditional tree structure yields no increase in accuracy. Second, building trees with data in whichthe class label and the attributes are independentoften results in large trees. In both cases, the pruning algorithms fail to control tree growth as one would expect themto. Weexplore this behaviortheoretically byconstructinga statistical modelof reduced error pruning. The model explains whyand whenthe pathologies occur, and makespredictions about howto lessen their effects. Thepredictions are operationalized in a variant of reducederror pruningthat is shownto control tree growthfar better than the original algorithm.

AAAI Conference 1996 Conference Paper

Searching for Planning Operators with Context-Dependent and Probabilistic Effects

• Tim Oates

Providing a complete and accurate domain model for an agent situated in a complex environment can be an extremely difficult task. Actions may have different effects depending on the context in which they are taken, and actions may or may not induce their intended effects, with the probability of success again depending on context. We present an algorithm for automatically learning planning operators with context-dependent and probabilistic effects in environments where exogenous events change the state of the world. Empirical results show that the algorithm successfully fh-rds operators that capture the true structure of an agent’ s interactions with its environment, and avoids spurious associations between actions and exogenous events.