Deeper neural networks are more difficult to train. We present a residual learning framework to ease the training of networks that are substantially deeper than those used previously. We explicitly reformulate the layers as learning residual functions with reference to the layer inputs, instead of learning unreferenced functions. We provide comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth. On the ImageNet dataset we evaluate residual nets with a depth of up to 152 layers—8× deeper than VGG nets [40] but still having lower complexity. An ensemble of these residual nets achieves 3.57% error on the ImageNet test set. This result won the 1st place on the ILSVRC 2015 classification task. We also present analysis on CIFAR-10 with 100 and 1000 layers. The depth of representations is of central importance for many visual recognition tasks. Solely due to our extremely deep representations, we obtain a 28% relative improvement on the COCO object detection dataset. Deep residual nets are foundations of our submissions to ILSVRC & COCO 2015 competitions1, where we also won the 1st places on the tasks of ImageNet detection, ImageNet localization, COCO detection, and COCO segmentation.

/pdf/deep-residual-learning-for-image-recognition-b75jg61qrq.pdf

Deep Residual Learning for Image Recognition

In this work we investigate the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting. Our main contribution is a thorough evaluation of networks of increasing depth using an architecture with very small (3x3) convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers. These findings were the basis of our ImageNet Challenge 2014 submission, where our team secured the first and the second places in the localisation and classification tracks respectively. We also show that our representations generalise well to other datasets, where they achieve state-of-the-art results. We have made our two best-performing ConvNet models publicly available to facilitate further research on the use of deep visual representations in computer vision.

Very Deep Convolutional Networks for Large-Scale Image Recognition

Deeper neural networks are more difficult to train. We present a residual learning framework to ease the training of networks that are substantially deeper than those used previously. We explicitly reformulate the layers as learning residual functions with reference to the layer inputs, instead of learning unreferenced functions. We provide comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth. On the ImageNet dataset we evaluate residual nets with a depth of up to 152 layers---8x deeper than VGG nets but still having lower complexity. An ensemble of these residual nets achieves 3.57% error on the ImageNet test set. This result won the 1st place on the ILSVRC 2015 classification task. We also present analysis on CIFAR-10 with 100 and 1000 layers. 
The depth of representations is of central importance for many visual recognition tasks. Solely due to our extremely deep representations, we obtain a 28% relative improvement on the COCO object detection dataset. Deep residual nets are foundations of our submissions to ILSVRC & COCO 2015 competitions, where we also won the 1st places on the tasks of ImageNet detection, ImageNet localization, COCO detection, and COCO segmentation.

Deep learning is a form of machine learning that enables computers to learn from experience and understand the world in terms of a hierarchy of concepts. Because the computer gathers knowledge from experience, there is no need for a human computer operator to formally specify all the knowledge that the computer needs. The hierarchy of concepts allows the computer to learn complicated concepts by building them out of simpler ones; a graph of these hierarchies would be many layers deep. This book introduces a broad range of topics in deep learning. The text offers mathematical and conceptual background, covering relevant concepts in linear algebra, probability theory and information theory, numerical computation, and machine learning. It describes deep learning techniques used by practitioners in industry, including deep feedforward networks, regularization, optimization algorithms, convolutional networks, sequence modeling, and practical methodology; and it surveys such applications as natural language processing, speech recognition, computer vision, online recommendation systems, bioinformatics, and videogames. Finally, the book offers research perspectives, covering such theoretical topics as linear factor models, autoencoders, representation learning, structured probabilistic models, Monte Carlo methods, the partition function, approximate inference, and deep generative models. Deep Learning can be used by undergraduate or graduate students planning careers in either industry or research, and by software engineers who want to begin using deep learning in their products or platforms. A website offers supplementary material for both readers and instructors.

Deep Learning

Today's conversational agents are restricted to simple standalone commands. In this paper, we present Iris, an agent that draws on human conversational strategies to combine commands, allowing it to perform more complex tasks that it has not been explicitly designed to support: for example, composing one command to "plot a histogram" with another to first "log-transform the data". To enable this complexity, we introduce a domain specific language that transforms commands into automata that Iris can compose, sequence, and execute dynamically by interacting with a user through natural language, as well as a conversational type system that manages what kinds of commands can be combined. We have designed Iris to help users with data science tasks, a domain that requires support for command combination. In evaluation, we find that data scientists complete a predictive modeling task significantly faster (2.6 times speedup) with Iris than a modern non-conversational programming environment. Iris supports the same kinds of commands as today's agents, but empowers users to weave together these commands to accomplish complex goals.

Iris: A Conversational Agent for Complex Tasks

Crowdsourcing systems accomplish large tasks with scale and speed by breaking work down into independent parts. However, many types of complex creative work, such as fiction writing, have remained out of reach for crowds because work is tightly interdependent: changing one part of a story may trigger changes to the overall plot and vice versa. Taking inspiration from how expert authors write, we propose a technique for achieving interdependent complex goals with crowds. With this technique, the crowd loops between reflection, to select a high-level goal, and revision, to decompose that goal into low-level, actionable tasks. We embody this approach in Mechanical Novel, a system that crowdsources short fiction stories on Amazon Mechanical Turk. In a field experiment, Mechanical Novel resulted in higher-quality stories than an iterative crowdsourcing workflow. Our findings suggest that orienting crowd work around high-level goals may enable workers to coordinate their effort to accomplish complex work.

Mechanical Novel: Crowdsourcing Complex Work through Reflection and Revision

Creating personal narratives helps people build meaning around their experiences. However, novices lack the knowledge and experience to create stories with strong narrative structure. Current storytelling tools often structure novice work through templates, enforcing a linear creative process that asks novices for materials they may not have. In this paper, we propose scaffolding creative work using storytelling patterns extracted from stories created by experts. Patterns are modular sets of related camera shots that expert videographers commonly use to achieve a specific narrative function. After identifying a set of patterns from high-quality storytelling videos, we created Motif, a mobile video storytelling application that allows users to construct video stories by combining these patterns. By making existing solutions used by experts available to novices, we encourage capturing shots with story structure and narrative goals in mind. In a controlled study where we asked participants to create travel video stories, videos created with patterns conveyed stronger narrative structure and were considered higher quality by expert evaluators than videos created without patterns.

/pdf/motif-supporting-novice-creativity-through-expert-patterns-24rycyti0f.pdf

Motif: Supporting Novice Creativity through Expert Patterns

Microblogs such as Twitter are a tremendous repository of user-generated content. Increasingly, we see tweets used as data sources for novel applications such as disaster mapping, brand sentiment analysis, and real-time visualizations. In each scenario, the workflow for processing tweets is ad-hoc, and a lot of unnecessary work goes into repeating common data processing patterns. We introduce TweeQL, a stream query processing language that presents a SQL-like query interface for unstructured tweets to generate structured data for downstream applications. We have built several tools on top of TweeQL, most notably TwitInfo, an event timeline generation and exploration interface that summarizes events as they are discussed on Twitter. Our demonstration will allow the audience to interact with both TweeQL and TwitInfo to convey the value of data embedded in tweets.

/pdf/tweets-as-data-demonstration-of-tweeql-and-twitinfo-44ceq1qt4w.pdf

Tweets as data: demonstration of TweeQL and Twitinfo

We have seen great progress in basic perceptual tasks such as object recognition and detection However, AI models still fail to match humans in high-level vision tasks due to the lack of capacities for deeper reasoning Recently the new task of visual question answering (QA) has been proposed to evaluate a model's capacity for deep image understanding Previous works have established a loose, global association between QA sentences and images However, many questions and answers, in practice, relate to local regions in the images We establish a semantic link between textual descriptions and image regions by object-level grounding It enables a new type of QA with visual answers, in addition to textual answers used in previous work We study the visual QA tasks in a grounded setting with a large collection of 7W multiple-choice QA pairs Furthermore, we evaluate human performance and several baseline models on the QA tasks Finally, we propose a novel LSTM model with spatial attention to tackle the 7W QA tasks

Michael S. Bernstein

Papers

Iris: A Conversational Agent for Complex Tasks

Mechanical Novel: Crowdsourcing Complex Work through Reflection and Revision

Motif: Supporting Novice Creativity through Expert Patterns

Tweets as data: demonstration of TweeQL and Twitinfo

Visual7W: Grounded Question Answering in Images