ML@GT Seminar Series

Series

ML@GT Seminar Series

Permanent Link

https://hdl.handle.net/1853/70935

Series Type

Event Series

Associated Organization(s)

Organizational Unit

Machine Learning Center

Organizational Unit

College of Computing

Full item page

Publication Search Results

Now showing 1 - 10 of 31

Learning to Optimize from Data: Faster, Better, and Guaranteed

( 2019-11-20) Wang, Zhangyang

Learning and optimization are closely related: state-of-the-art learning problems hinge on the sophisticated design of optimizers. On the other hand, the optimization cannot be considered as independent from data, since data may implicitly contain important information that guides optimization, as seen in the recent waves of meta-learning or learning to optimize. This talk will discuss Learning Augmented Optimization (LAO), a nascent area that bridges classical optimization with the latest data-driven learning, by augmenting classical model-based optimization with learning-based components. By adapting their behavior to the properties of the input distribution, the ``augmented'' algorithms may reduce their complexities by magnitudes, and/or improve their accuracy, while still preserving favorable theoretical guarantees such as convergence. I will start by diving into a case study on exploiting deep learning to solve the convex LASSO problem, showing its linear convergence in addition to superior parameter efficiency. Then, our discussions will be extended to applying LAO approaches to solving plug-and-play (PnP) optimization, and population-based optimization. I will next demonstrate our recent results on ensuring the robustness of LAO, say how applicable the algorithm remains to be, if the testing problem instances deviate from the training problem distribution. The talk will be concluded by a few thoughts and reflections, as well as pointers to potential future directions.
The Data-Driven Analysis of Literature

( 2019-11-15) Bamman, David

Literary novels push the limits of natural language processing. While much work in NLP has been heavily optimized toward the narrow domains of news and Wikipedia, literary novels are an entirely different animal--the long, complex sentences in novels strain the limits of syntactic parsers with super-linear computational complexity, their use of figurative language challenges representations of meaning based on neo-Davidsonian semantics, and their long length (ca. 100,000 words on average) rules out existing solutions for problems like coreference resolution that expect a small set of candidate antecedents. At the same time, fiction drives computational research questions that are uniquely interesting to that domain. In this talk, I'll outline some of the opportunities that NLP presents for research in the quantitative analysis of culture--including measuring the disparity in attention given to characters as a function of their gender over two hundred years of literary history (Underwood et al. 2018)--and describe our progress to date on two problems essential to a more complex representation of plot: recognizing the entities in literary texts, such as the characters, locations, and spaces of interest (Bamman et al. 2019) and identifying the events that are depicted as having transpired (Sims et al. 2019). Both efforts involve the creation of a new dataset of 200,000 words evenly drawn from 100 different English-language literary texts and building computational models to automatically identify each phenomenon. This is joint work with Matt Sims, Ted Underwood, Sabrina Lee, Jerry Park, Sejal Popat and Sheng Shen.
A Discussion on Fairness in Machine Learning with Georgia Tech Faculty

( 2019-11-06) Cummings, Rachel ; Desai, Devan ; Gupta, Swati ; Hoffman, Judy

Fairness in machine learning and artificial intelligence is a hot, and important topic in tech today. Join Georgia Tech faculty members Judy Hoffman, Rachel Cummings, Deven Desai, and Swati Gupta for a panel discussion on their work in regards to fairness and their motivations behind it. Sponsored by the Machine Learning Center at Georgia Tech.
NLP Approaches to Campaign Classification

( 2019-10-17) Ahmed, Muhammed

Mailchimp is the world's largest marketing automation platform. Over a billion emails are sent by it every day, which raises the question: what exactly are users sending? We'll do a deep dive into the natural language processing techniques utilized by Mailchimp to make sense of users' content and classify campaigns, whether it's for predicting a customer's business vertical, or for preventing those with malicious intent from using the platform.
The Geometry of Community Detection via the MMSE Matrix

( 2019-09-04) Reeves, Galen

The information-theoretic limits of community detection have been studied extensively for network models with high levels of symmetry or homogeneity. In this talk, Reeves will present a new approach that applies to a broader class of network models that allow for variability in the sizes and behaviors of the different communities, and thus better reflect the behaviors observed in real-world networks. The results show that the ability to detect communities can be described succinctly in terms of a matrix of effective signal-to-noise ratios that provides a geometrical representation of the relationships between the different communities. This characterization follows from a matrix version of the I-MMSE relationship and generalizes the concept of an effective scalar signal-to-noise ratio introduced in previous work. This work can be found online at https://arxiv.org/abs/1907.02496
Perception at Magic Leap

( 2019-04-19) Swaminathan, Ashwin

This talk presents the importance of Computer Vision and Deep learning techniques in making Magic Leap an effective spatial computing platform. The four fundamental modalities are introduced: head pose tracking, world reconstruction, eye tracking, and hand tracking; emphasizing on the two main general themes: Understanding the world (spatial localization, environment mapping) and Understanding user’s intent (eye, gaze, and hands). The talk will provide a deep dive into the main modalities along with key challenges and open problems. We will go over some technical challenges that our team of researchers and engineers are tacking to make computer vision work for see-through wearable Mixed Reality device. We will start with an overview of the past 5 years of development in inside-out pose estimation (Visual Inertial Odometry), scene reconstruction, eye tracking, and sensor calibration and then dive into the long list of technical challenges that are left to solve.
AI Driven Design Approach

( 2019-04-03) Srivastava, Sanjeev

Design Space Exploration (DSE) is an activity that is performed to systematically analyze several design points and then select the design(s) based on parameters of interest and design requirements. For complex systems, design engineers spend a lot of time generating new design iterations using simulation-based tools. However, they do not explore the complete design space due to time-consuming design creation and analyses process, and hence do not generate optimal design but rather settle for a sub-optimal design that meets the requirements. Recently, there have been advances in the areas of Artificial Intelligence (AI), cognitive computing, Internet of Things, 3D printing, advanced robotics, that provide capabilities to transform the current design process by automating various design steps, speeding up the design and analysis toolchain, and creative decision making. Within Siemens CT we have been developing a Generative Design framework, which utilizes the AI and knowledge representation techniques in combination with traditional design methods to generate better designs in a faster manner. The talk will address various technical areas of this framework and the various AI methods which we are currently applying to optimally design parts, systems, and processes.
Situated Natural Language Understanding

( 2019-03-08) Misra, Dipendra
Contextual AI - The Next Frontier Towards Human-Centric Artificial Intelligence

( 2019-02-28) Brdiczka, Oliver

This talk motivates a more human-centric wave of AI, dubbed Contextual AI. Contextual AI does not refer to a specific algorithm or machine learning method - instead, it takes a human-centric view and approach to AI. The core is the definition of a set of requirements that enable a symbiotic relationship between AI and humans. Adobe is at the forefront of this new wave enabling creatives to leverage and interact with AI more naturally during the human creative process. We will illustrate what Adobe is doing in the space with a number of examples, including deep learning for content understanding and neural stylization, and demos of Adobe’s Creative Assistant.
Automated Perception in the Real World: The Problem of Scarce Data

( 2018-11-30) Ernst, Jan

Machine perception is a key step toward artificial intelligence in domains such as self-driving cars, industrial automation, and robotics. Much progress has been made in the past decade, driven by machine learning, ever-increasing computational power, and the reliance on (seemingly) vast data sets. There are however critical issues in translating academic progress into the real world: available data sets may not match real-world environments well, and even if they are abundant and matching well, then interesting samples from a real-world perspective may be exceedingly rare and thus still be too sparsely represented to learn from directly. In this talk, I illustrate how we have approached this problem strategically as an example of industrial R&D from inception to product. I will also go in-depth on an approach to automatically infer previously unseen data by learning compositional visual concepts via mutual cycle consistency.