Search Results for author: John Canny

Found 40 papers, 17 papers with code

A Human-Inspired Reading Agent with Gist Memory of Very Long Contexts

no code implementations15 Feb 2024 Kuang-Huei Lee, Xinyun Chen, Hiroki Furuta, John Canny, Ian Fischer

Current Large Language Models (LLMs) are not only limited to some maximum context length, but also are not able to robustly consume long inputs.

Reading Comprehension Retrieval

Moral Foundations of Large Language Models

1 code implementation23 Oct 2023 Marwa Abdulhai, Gregory Serapio-Garcia, Clément Crepy, Daria Valter, John Canny, Natasha Jaques

Finally, we show that we can adversarially select prompts that encourage the moral to exhibit a particular set of moral foundations, and that this can affect the model's behavior on downstream tasks.

IC3: Image Captioning by Committee Consensus

1 code implementation2 Feb 2023 David M. Chan, Austin Myers, Sudheendra Vijayanarasimhan, David A. Ross, John Canny

If you ask a human to describe an image, they might do so in a thousand different ways.

Image Captioning

Distribution Aware Metrics for Conditional Natural Language Generation

no code implementations15 Sep 2022 David M Chan, Yiming Ni, David A Ross, Sudheendra Vijayanarasimhan, Austin Myers, John Canny

In this work we argue that existing metrics are not appropriate for domains such as visual description or summarization where ground truths are semantically diverse, and where the diversity in those captions captures useful additional information about the context.

speech-recognition Speech Recognition +1

LAVA: Language Audio Vision Alignment for Contrastive Video Pre-Training

no code implementations16 Jul 2022 Sumanth Gurram, Andy Fang, David Chan, John Canny

Generating representations of video data is of key importance in advancing the field of machine perception.

Action Recognition Contrastive Learning +1

An Embedding-Dynamic Approach to Self-supervised Learning

no code implementations7 Jul 2022 Suhong Moon, Domas Buracas, Seunghyun Park, Jinkyu Kim, John Canny

It also uses a purely-dynamic local dispersive force (Brownian motion) that shows improved performance over other methods and does not require knowledge of other particle coordinates.

Classification Image Classification +7

Towards Understanding How Machines Can Learn Causal Overhypotheses

1 code implementation16 Jun 2022 Eliza Kosoy, David M. Chan, Adrian Liu, Jasmine Collins, Bryanna Kaufmann, Sandy Han Huang, Jessica B. Hamrick, John Canny, Nan Rosemary Ke, Alison Gopnik

Recent work in machine learning and cognitive science has suggested that understanding causal information is essential to the development of intelligence.

BIG-bench Machine Learning Causal Inference

NewsPod: Automatic and Interactive News Podcasts

no code implementations15 Feb 2022 Philippe Laban, Elicia Ye, Srujay Korlakunta, John Canny, Marti A. Hearst

News podcasts are a popular medium to stay informed and dive deep into news topics.

Misinformation Detection in Social Media Video Posts

no code implementations15 Feb 2022 Kehan Wang, David Chan, Seth Z. Zhao, John Canny, Avideh Zakhor

With the growing adoption of short-form video by social media platforms, reducing the spread of misinformation through video posts has become a critical challenge for social media providers.

Contrastive Learning Language Modelling +3

Sketch-based Creativity Support Tools using Deep Learning

no code implementations19 Nov 2021 Forrest Huang, Eldon Schoop, David Ha, Jeffrey Nichols, John Canny

Sketching is a natural and effective visual communication medium commonly used in creative processes.

Retrieval

Compressive Visual Representations

1 code implementation NeurIPS 2021 Kuang-Huei Lee, Anurag Arnab, Sergio Guadarrama, John Canny, Ian Fischer

We verify this by developing SimCLR and BYOL formulations compatible with the Conditional Entropy Bottleneck (CEB) objective, allowing us to both measure and control the amount of compression in the learned representation, and observe their impact on downstream tasks.

Contrastive Learning Self-Supervised Image Classification

DCUR: Data Curriculum for Teaching via Samples with Reinforcement Learning

1 code implementation15 Sep 2021 Daniel Seita, Abhinav Gopal, Zhao Mandi, John Canny

Then, students learn by running either offline RL or by using teacher data in combination with a small amount of self-generated data.

Offline RL reinforcement-learning +1

What's The Latest? A Question-driven News Chatbot

no code implementations ACL 2020 Philippe Laban, John Canny, Marti A. Hearst

This work describes an automatic news chatbot that draws content from a diverse set of news articles and creates conversations with a user about the news.

Chatbot

The Summary Loop: Learning to Write Abstractive Summaries Without Examples

1 code implementation ACL 2020 Philippe Laban, Andrew Hsi, John Canny, Marti A. Hearst

This work presents a new approach to unsupervised abstractive summarization based on maximizing a combination of coverage and fluency for a given length constraint.

Abstractive Text Summarization News Summarization

Active Learning for Video Description With Cluster-Regularized Ensemble Ranking

no code implementations27 Jul 2020 David M. Chan, Sudheendra Vijayanarasimhan, David A. Ross, John Canny

Automatic video captioning aims to train models to generate text descriptions for all segments in a video, however, the most effective approaches require large amounts of manual annotation which is slow and expensive.

Active Learning Video Captioning +1

A Dataset and Benchmarks for Multimedia Social Analysis

no code implementations5 Jun 2020 Bofan Xue, David Chan, John Canny

We present a new publicly available dataset with the goal of advancing multi-modality learning by offering vision and language data within the same context.

Image Captioning Image Classification +2

Scones: Towards Conversational Authoring of Sketches

no code implementations12 May 2020 Forrest Huang, Eldon Schoop, David Ha, John Canny

Iteratively refining and critiquing sketches are crucial steps to developing effective designs.

Exploring Exploration: Comparing Children with RL Agents in Unified Environments

1 code implementation6 May 2020 Eliza Kosoy, Jasmine Collins, David M. Chan, Sandy Huang, Deepak Pathak, Pulkit Agrawal, John Canny, Alison Gopnik, Jessica B. Hamrick

Research in developmental psychology consistently shows that children explore the world thoroughly and efficiently and that this exploration allows them to learn.

Measuring the Reliability of Reinforcement Learning Algorithms

1 code implementation ICLR 2020 Stephanie C. Y. Chan, Samuel Fishman, John Canny, Anoop Korattikara, Sergio Guadarrama

To aid RL researchers and production users with the evaluation and improvement of reliability, we propose a set of metrics that quantitatively measure different aspects of reliability.

reinforcement-learning Reinforcement Learning (RL)

Grounding Human-to-Vehicle Advice for Self-driving Vehicles

no code implementations CVPR 2019 Jinkyu Kim, Teruhisa Misu, Yi-Ting Chen, Ashish Tawari, John Canny

We show that taking advice improves the performance of the end-to-end network, while the network cues on a variety of visual features that are provided by advice.

ZPD Teaching Strategies for Deep Reinforcement Learning from Demonstrations

2 code implementations26 Oct 2019 Daniel Seita, David Chan, Roshan Rao, Chen Tang, Mandi Zhao, John Canny

Learning from demonstrations is a popular tool for accelerating and reducing the exploration requirements of reinforcement learning.

Atari Games Q-Learning +2

Deep Imitation Learning of Sequential Fabric Smoothing From an Algorithmic Supervisor

1 code implementation23 Sep 2019 Daniel Seita, Aditya Ganapathi, Ryan Hoque, Minho Hwang, Edward Cen, Ajay Kumar Tanwani, Ashwin Balakrishna, Brijen Thananjeyan, Jeffrey Ichnowski, Nawid Jamali, Katsu Yamane, Soshi Iba, John Canny, Ken Goldberg

In 180 physical experiments with the da Vinci Research Kit (dVRK) surgical robot, RGBD policies trained in simulation attain coverage of 83% to 95% depending on difficulty tier, suggesting that effective fabric smoothing policies can be learned from an algorithmic supervisor and that depth sensing is a valuable addition to color alone.

Imitation Learning

Evaluating Protein Transfer Learning with TAPE

5 code implementations NeurIPS 2019 Roshan Rao, Nicholas Bhattacharya, Neil Thomas, Yan Duan, Xi Chen, John Canny, Pieter Abbeel, Yun S. Song

Semi-supervised learning has emerged as an important paradigm in protein modeling due to the high cost of acquiring supervised protein labels, but the current literature is fragmented when it comes to datasets and standardized evaluation techniques.

BIG-bench Machine Learning Representation Learning +1

Risk Averse Robust Adversarial Reinforcement Learning

no code implementations31 Mar 2019 Xinlei Pan, Daniel Seita, Yang Gao, John Canny

In this paper we introduce risk-averse robust adversarial reinforcement learning (RARARL), using a risk-averse protagonist and a risk-seeking adversary.

reinforcement-learning Reinforcement Learning (RL)

Periphery-Fovea Multi-Resolution Driving Model guided by Human Attention

1 code implementation24 Mar 2019 Ye Xia, Jinkyu Kim, John Canny, Karl Zipser, David Whitney

Inspired by human vision, we propose a new periphery-fovea multi-resolution driving model that predicts vehicle speed from dash camera videos.

Label and Sample: Efficient Training of Vehicle Object Detector from Sparsely Labeled Data

no code implementations26 Aug 2018 Xinlei Pan, Sung-Li Chiang, John Canny

First, we note that an optimal subset (relative to all the objects encountered and labeled) of labeled objects in images can be obtained by importance sampling using gradients of the recognition network.

object-detection Object Detection +1

Textual Explanations for Self-Driving Vehicles

2 code implementations ECCV 2018 Jinkyu Kim, Anna Rohrbach, Trevor Darrell, John Canny, Zeynep Akata

Finally, we explore a version of our model that generates rationalizations, and compare with introspective explanations on the same video segments.

Fast and Reliable Autonomous Surgical Debridement with Cable-Driven Robots Using a Two-Phase Calibration Procedure

1 code implementation19 Sep 2017 Daniel Seita, Sanjay Krishnan, Roy Fox, Stephen McKinley, John Canny, Ken Goldberg

In Phase II (fine), the bias from Phase I is applied to move the end-effector toward a small set of specific target points on a printed sheet.

Robotics

General models for rational cameras and the case of two-slit projections

no code implementations CVPR 2017 Matthew Trager, Bernd Sturmfels, John Canny, Martial Hebert, Jean Ponce

The rational camera model recently introduced in [19] provides a general methodology for studying abstract nonlinear imaging systems and their multi-view geometry.

An Efficient Minibatch Acceptance Test for Metropolis-Hastings

no code implementations19 Oct 2016 Daniel Seita, Xinlei Pan, Haoyu Chen, John Canny

We present a novel Metropolis-Hastings method for large datasets that uses small expected-size minibatches of data.

Fast Parallel SAME Gibbs Sampling on General Discrete Bayesian Networks

no code implementations19 Nov 2015 Daniel Seita, Haoyu Chen, John Canny

A fundamental task in machine learning and related fields is to perform inference on Bayesian networks.

SAME but Different: Fast and High-Quality Gibbs Parameter Estimation

2 code implementations18 Sep 2014 Huasha Zhao, Biye Jiang, John Canny

SAME (State Augmentation for Marginal Estimation) \cite{Doucet99, Doucet02} is an approach to MAP parameter estimation which gives improved parameter estimates over direct Gibbs sampling.

Bayesian Inference Vocal Bursts Intensity Prediction

Sparse Allreduce: Efficient Scalable Communication for Power-Law Data

no code implementations11 Dec 2013 Huasha Zhao, John Canny

Many large datasets exhibit power-law statistics: The web graph, social networks, text data, click through data etc.

Clustering Topic Models

Factor Modeling for Advertisement Targeting

no code implementations NeurIPS 2009 Ye Chen, Michael Kapralov, John Canny, Dmitry Y. Pavlov

We adapt a probabilistic latent variable model, namely GaP (Gamma-Poisson), to ad targeting in the contexts of sponsored search (SS) and behaviorally targeted (BT) display advertising.

Cannot find the paper you are looking for? You can Submit a new open access paper.