Browsing by Author "Jaakkola, Tommi S"
Now showing items 1-20 of 42
-
Active boundary annotation using random MAP perturbations
(PLMR, 2014-04)We address the problem of efficiently annotating labels of objects when they are structured. Often the distribution over labels can be described using a joint potential function over the labels for which sampling is provably ... -
Approximate inference in additive factorial HMMs with application to energy disaggregation
(Proceedings of Machine Learning Research, 2012-04)This paper considers additive factorial hidden Markov models, an extension to HMMs where the state factors into multiple independent chains, and the output is an additive function of all the hidden states. Although such ... -
Bidirectional Inference Networks:A Class of Deep Bayesian Networks for Health Profiling
(Association for the Advancement of Artificial Intelligence (AAAI), 2019-01)We consider the problem of inferring the values of an arbitrary set of variables (e.g., risk of diseases) given other observed variables (e.g., symptoms and diagnosed diseases) and high-dimensional signals (e.g., MRI images ... -
Collaborative future event recommendation
(Association for Computing Machinery, 2010-10)We demonstrate a method for collaborative ranking of future events. Previous work on recommender systems typically relies on feedback on a particular item, such as a movie, and generalizes this to other items or other ... -
Controlling privacy in recommender systems
(Neural Information Processing Systems, 2014)Recommender systems involve an inherent trade-off between accuracy of recommendations and the extent to which users are willing to release information about their preferences. In this paper, we explore a two-tiered notion ... -
Convolutional Embedding of Attributed Molecular Graphs for Physical Property Prediction
(American Chemical Society (ACS), 2017-07)The task of learning an expressive molecular representation is central to developing quantitative structure–activity and property relationships. Traditional approaches rely on group additivity rules, empirical measurements ... -
Current and Future Roles of Artificial Intelligence in Medicinal Chemistry Synthesis
(American Chemical Society (ACS), 2020-04)Artificial intelligence and machine learning have demonstrated their potential role in predictive chemistry and synthetic planning of small molecules; there are at least a few reports of companies employing in silico ... -
Deep learning identifies synergistic drug combinations for treating COVID-19
(National Academy of Sciences, 2021-09)Effective treatments for COVID-19 are urgently needed. However, discovering single-agent therapies with activity against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has been challenging. Combination therapies ... -
Deriving neural architectures from sequence and graph kernels
(MLResearch Press, 2017)The design of neural architectures for structured objects is typically guided by experimental insights rather than a formal process. In this work, we appeal to kernels over combinatorial structures, such as sequences and ... -
Direct optimization through arg max for discrete variational auto-encoder
(Morgan Kaufmann Publishers, 2019-12)Reparameterization of variational auto-encoders with continuous random variables is an effective method for reducing the variance of their gradient estimates. In the discrete case, one can perform reparametrization using ... -
Dual decomposition for parsing with non-projective head automata
(Association for Computational Linguistics, 2010-10)This paper introduces algorithms for non-projective parsing based on dual decomposition. We focus on parsing algorithms for non-projective head automata, a generalization of head-automata models to non-projective structures. ... -
From random walks to distances on unweighted graphs
(Neural Information Processing Systems Foundation, Inc., 2015-12)Large unweighted directed graphs are commonly used to capture relations between entities. A fundamental problem in the analysis of such networks is to properly define the similarity or dissimilarity between any two vertices. ... -
A game theoretic approach to class-wise selective rationalization
(Neural Information Processing Systems Foundation, Inc., 2019)Selection of input features such as relevant pieces of text has become a common technique of highlighting how complex neural predictors operate. The selection can be optimized post-hoc for trained models or incorporated ... -
Generative models for graph-based protein design
(Neural Information Processing Systems Foundation, Inc., 2019)Engineered proteins offer the potential to solve many problems in biomedicine, energy, and materials science, but creating designs that succeed is difficult in practice. A significant aspect of this challenge is the complex ... -
Greed Is Good If Randomized: New Inference for Dependency Parsing
(2014-10)Dependency parsing with high-order features results in a provably hard decoding problem. A lot of work has gone into developing powerful optimization methods for solving these combinatorial problems. In contrast, we explore, ... -
Grounding Language for Transfer in Deep Reinforcement Learning
(AI Access Foundation, 2018)© 2018 AI Access Foundation. All rights reserved. In this paper, we explore the utilization of natural language to drive transfer for reinforcement learning (RL). Despite the wide-spread application of deep RL techniques, ... -
High Dimensional Inference with Random Maximum A-Posteriori Perturbations
(Institute of Electrical and Electronics Engineers (IEEE), 2019-05)This paper presents a new approach, called perturb-max, for high-dimensional statistical inference in graphical models that is based on applying random perturbations followed by optimization. This framework injects randomness ... -
Inverse Covariance Estimation for High-Dimensional Data in Linear Time and Space: Spectral Methods for Riccati and Sparse Models
(Association for Uncertainty in Artificial Intelligence (AUAI), 2013-07)We propose maximum likelihood estimation for learning Gaussian graphical models with a Gaussian (ℓ[2 over 2]) prior on the parameters. This is in contrast to the commonly used Laplace (ℓ[subscript 1) prior for encouraging ... -
Learning bayesian network structure using lp relaxations
(Society for Artificial Intelligence and Statistics, 2010-05)We propose to solve the combinatorial problem of finding the highest scoring Bayesian network structure from data. This structure learning problem can be viewed as an inference problem where the variables specify ... -
Learning efficient random maximum a-posteriori predictors with non-decomposable loss functions
(Neural Information Processing Systems, 2013)In this work we develop efficient methods for learning random MAP predictors for structured label problems. In particular, we construct posterior distributions over perturbations that can be adjusted via stochastic gradient ...
