VideosSee AI2's full collection of videos on our YouTube channel.
Viewing 141-150 of 164 videos
- April 7, 2015 | Dani YogatamaThe majority of NLP research focuses on improving NLP systems by designing better model classes (e.g., non-linear models, latent variable models). In this talk, I will describe a complementary approach based on incorporation of linguistic bias and optimization of text representations that is applicable to several model classes. First, I will present a structured regularizer that is suitable for the problem when only some parts of an input are relevant to the prediction task (e.g., sentences in text, entities in scenes of images) and an efficient algorithm based on the alternating direction method of multipliers to solve the resulting optimization problem. I will then show how such regularizer can be used to incorporate linguistic structures into a text classification model. In the second part of the talk, I will present our first step towards building a black box NLP system that automatically chooses the best text representation for a given dataset by treating it as a global optimization problem. I will also briefly describe an improved algorithm that can generalize across multiple datasets for faster optimization. I will conclude by discussing how such a framework can be applied to other NLP problems.
- March 31, 2015 |In many real-world applications of AI and machine learning, such as natural language processing, computer vision and knowledge base construction, data sources possess a natural internal structure, which can be exploited to improve predictive accuracy. Sometimes the structure can be very large, containing many interdependent inputs and outputs. Learning from data with large internal structure poses many compelling challenges, one of which is that fully-labeled examples (required for supervised learning) are difficult to acquire. This is especially true in applications like image segmentation, annotating video data, and knowledge base construction.
- March 27, 2015 | Sonal GuptaAlthough most work in information extraction (IE) focuses on tasks that have abundant training data, in practice, many IE problems do not have any supervised training data. State-of-the-art supervised techniques like conditional random fields are impractical for such real world applications because: (1) they require large and expensive labeled corpora; (2) it is difficult to interpret them and analyze errors, an often-ignored but important feature; and (3) they are hard to calibrate, for example, to reliably extract only high-precision extractions.
- March 17, 2015 | Congle ZhangMost approaches to relation extraction, the task of extracting ground facts from natural language text, are based on machine learning and thus starved by scarce training data. Manual annotation is too expensive to scale to a comprehensive set of relations. Distant supervision, which automatically creates training data, only works with relations that already populate a knowledge base (KB). Unfortunately, KBs such as FreeBase rarely cover event relations (e.g. “person travels to location”). Thus, the problem of extracting a wide range of events — e.g., from news streams — is an important, open challenge.
- March 12, 2015 | Vicente OrdonezRecently, there has been great progress in both computer vision and natural language processing in representing and recognizing semantic units like objects, attributes, named entities, or constituents. These advances provide opportunities to create systems able to interpret and describe the visual world using natural language. This is in contrast to traditional computer vision systems, which typically output a set of disconnected labels, object locations, or annotations for every pixel in an image. The rich visually descriptive language produced by people incorporates world knowledge and human intuition that often can not be captured by other types of annotations. In this talk, I will present several approaches that explore the connections between language, perception, and vision at three levels: learning how to name objects, generating referring expressions for objects in natural scenes, and producing general image descriptions. These methods provide a framework to augment computer vision systems with linguistic information and to take advantage of the vast amount of text associated with images on the web. I will also discuss some of the intuitions from linguistics and perception behind these efforts and how they potentially connect to the larger goal of creating visual systems that can better learn from and communicate with people.
- March 11, 2015 | Joel PfeifferNetworks provide an effective representation to model many real-world domains, with edges (e.g., friendships, citations, hyperlinks) representing relationships between items (e.g., individuals, papers, webpages). By understanding common network features, we can develop models of the distribution from which the network was likely sampled. These models can be incorporated into real world tasks, such as modeling partially observed networks for improving relational machine learning, performing hypothesis tests for anomaly detection, or simulating algorithms on large scale (or future) datasets. However, naively sampling networks does not scale to real-world domains; for example, drawing a single random network sample consisting of a billion users would take approximately a decade with modern hardware.
- March 3, 2015 | Ankur ParikhBeing able to effectively model latent structure in data is a key challenge in modern AI research, particularly in Natural Language Processing (NLP) where it is crucial to discover and leverage syntactic and semantic relationships that may not be explicitly annotated in the training set. Unfortunately, while incorporating latent variables to represent hidden structure can substantially increase representation power, the key problems of model design and learning become significantly more complicated. For example, unlike fully observed models, latent variable models can suffer from non-identifiability, making it difficult to distinguish the desired latent structure from the others. Moreover, learning is usually formulated as a non-convex optimization problem, leading to the use of local search heuristics that may become trapped in local optima.
- February 26, 2015 | Ken ForbusCreating systems that can work with people, using natural modalities, as apprentices is a key step towards human-level AI. This talk will describe how my group is combining research on sketch understanding, natural language understanding, and analogical learning within the Companion cognitive architecture to create systems that can reason and learn about science by working with people. Some promising results will be described (e.g. solving conceptual physics problems involving sketches, modeling conceptual change, learning by reading) as well as work in progress (e.g. interactive knowledge capture via analogy).
- February 5, 2015 | Bhavana DalviSemi-supervised learning (SSL) has been widely used over a decade for various tasks -- including knowledge acquisition-- that lack large amount of training data. My research proposes a novel learning scenario in which the system knows a few categories in advance, but the rest of the categories are unanticipated and need to be discovered from the unlabeled data. With the availability of enormous unlabeled datasets at low cost, and difficulty of collecting labeled data for all possible categories, it becomes even more important to adapt traditional semi-supervised learning techniques to such realistic settings.
- January 7, 2015 | Been KimI will present the Bayesian Case Model (BCM), a general framework for Bayesian case-based reasoning (CBR) and prototype classification and clustering. BCM brings the intuitive power of CBR to a Bayesian generative framework. The BCM learns prototypes, the ``quintessential" observations that best represent clusters in a data set, by performing joint inference on cluster labels, prototypes and important features. Simultaneously, BCM pursues sparsity by learning subspaces, the sets of features that play important roles in the characterization of the prototypes. The prototype and subspace representation provides quantitative benefits in interpretability while preserving classification accuracy. Human subject experiments verify statistically significant improvements to participants' understanding when using explanations produced by BCM, compared to those given by prior art.