no code implementations • ICML 2020 • Vikas Garg, Tommi Jaakkola
Our games take as input, e. g., UN resolution to be voted on, and map such contexts to initial strategies, player utilities, and interactions.
no code implementations • ICML 2020 • Vikas Garg, Tommi Jaakkola
Our games take as input, e. g., UN resolution to be voted on, and map such contexts to initial strategies, player utilities, and interactions.
no code implementations • ICML 2020 • Wengong Jin, Regina Barzilay, Tommi Jaakkola
These rationales are identified from molecules as substructures that are likely responsible for each property of interest.
1 code implementation • 1 Feb 2023 • Yilun Xu, Shangyuan Tong, Tommi Jaakkola
We show that the procedure indeed helps in the challenging intermediate regime by reducing (the trace of) the covariance of training targets.
no code implementations • 28 Nov 2022 • Anurag Ajay, Yilun Du, Abhi Gupta, Joshua Tenenbaum, Tommi Jaakkola, Pulkit Agrawal
We further demonstrate the advantages of modeling policies as conditional diffusion models by considering two other conditioning variables: constraints and skills.
no code implementations • 14 Oct 2022 • Bracha Laufer-Goldshtein, Adam Fisch, Regina Barzilay, Tommi Jaakkola
Machine learning applications frequently come with multiple diverse objectives and constraints that can change over time.
1 code implementation • 13 Oct 2022 • Xiang Fu, Zhenghao Wu, Wujie Wang, Tian Xie, Sinan Keten, Rafael Gomez-Bombarelli, Tommi Jaakkola
Our benchmark suite comes with a comprehensive open-source codebase for training and simulation with ML FFs to facilitate further work.
1 code implementation • 4 Oct 2022 • Gabriele Corso, Hannes Stärk, Bowen Jing, Regina Barzilay, Tommi Jaakkola
Predicting the binding structure of a small molecule ligand to a protein -- a task known as molecular docking -- is critical to drug design.
Ranked #1 on
Blind Docking
on PDBbind
1 code implementation • 22 Sep 2022 • Yilun Xu, Ziming Liu, Max Tegmark, Tommi Jaakkola
We interpret the data points as electrical charges on the $z=0$ hyperplane in a space augmented with an additional dimension $z$, generating a high-dimensional electric field (the gradient of the solution to Poisson equation).
Ranked #15 on
Image Generation
on CIFAR-10
no code implementations • 25 Aug 2022 • Adam Fisch, Tommi Jaakkola, Regina Barzilay
Providing calibrated uncertainty estimates alongside predictions -- probabilities that correspond to true frequencies -- can be as important as having predictions that are simply accurate on average.
no code implementations • 14 Jul 2022 • Wengong Jin, Regina Barzilay, Tommi Jaakkola
The binding affinity is governed by the 3D binding interface where antibody residues (paratope) closely interact with antigen residues (epitope).
no code implementations • 8 Jun 2022 • Brian L. Trippe, Jason Yim, Doug Tischer, Tamara Broderick, David Baker, Regina Barzilay, Tommi Jaakkola
Construction of a scaffold structure that supports a desired motif, conferring protein function, shows promise for the design of vaccines and enzymes.
1 code implementation • 1 Jun 2022 • Bowen Jing, Gabriele Corso, Jeffrey Chang, Regina Barzilay, Tommi Jaakkola
Molecular conformer generation is a fundamental task in computational chemistry.
1 code implementation • 3 May 2022 • Bowen Jing, Gabriele Corso, Renato Berlinghieri, Tommi Jaakkola
Score-based models generate samples by mapping noise to data (and vice versa) via a high-dimensional diffusion process.
Ranked #5 on
Image Generation
on CIFAR-10
1 code implementation • 21 Apr 2022 • Xiang Fu, Tian Xie, Nathan J. Rebello, Bradley D. Olsen, Tommi Jaakkola
Molecular dynamics (MD) simulation is the workhorse of various scientific domains but is limited by high computational cost.
1 code implementation • 15 Feb 2022 • Adam Fisch, Tal Schuster, Tommi Jaakkola, Regina Barzilay
We propose to trade coverage for a notion of precision by enforcing that the presence of incorrect candidates in the predicted conformal sets (i. e., the total number of false positives) is bounded according to a user-specified tolerance.
1 code implementation • 7 Feb 2022 • Hannes Stärk, Octavian-Eugen Ganea, Lagnajit Pattanaik, Regina Barzilay, Tommi Jaakkola
Predicting how a drug-like molecule binds to a specific protein target is a core problem in drug discovery.
Ranked #5 on
Blind Docking
on PDBBind
1 code implementation • ICLR 2022 • Yilun Xu, Hao He, Tianxiao Shen, Tommi Jaakkola
We propose to identify directions invariant to a given classifier so that these directions can be controlled in tasks such as style transfer.
1 code implementation • ICLR 2022 • Octavian-Eugen Ganea, Xinyuan Huang, Charlotte Bunne, Yatao Bian, Regina Barzilay, Tommi Jaakkola, Andreas Krause
Protein complex formation is a central problem in biology, being involved in most of the cell's processes, and essential for applications, e. g. drug design or protein engineering.
no code implementations • NeurIPS Workshop AI4Scien 2021 • Benson Chen, Xiang Fu, Regina Barzilay, Tommi Jaakkola
Equipped with the learned fragment vocabulary, we propose Fragment-based Sequential Translation (FaST), which learns a reinforcement learning (RL) policy to iteratively translate model-discovered molecules into increasingly novel molecules while satisfying desired properties.
2 code implementations • 19 Oct 2021 • Yilun Xu, Tommi Jaakkola
We further demonstrate the impact of optimizing such transfer risk on two controlled settings, each representing a different pattern of environment shift, as well as on two real-world datasets.
1 code implementation • ICLR 2022 • Tian Xie, Xiang Fu, Octavian-Eugen Ganea, Regina Barzilay, Tommi Jaakkola
Generating the periodic structure of stable materials is a long-standing challenge for the material design community.
1 code implementation • ICLR 2022 • Wengong Jin, Jeremy Wohlwend, Regina Barzilay, Tommi Jaakkola
In this paper, we propose a generative model to automatically design the CDRs of antibodies with enhanced binding specificity or neutralization capabilities.
1 code implementation • 29 Jun 2021 • Xiang Fu, Ge Yang, Pulkit Agrawal, Tommi Jaakkola
Current model-based reinforcement learning methods struggle when operating from complex visual scenes due to their inability to prioritize task-relevant features.
Model-based Reinforcement Learning
reinforcement-learning
+1
no code implementations • CVPR 2021 • Karren Yang, Samuel Goldman, Wengong Jin, Alex X. Lu, Regina Barzilay, Tommi Jaakkola, Caroline Uhler
In this paper, we aim to synthesize cell microscopy images under different molecular interventions, motivated by practical applications to drug development.
no code implementations • EMNLP 2021 • Tal Schuster, Adam Fisch, Tommi Jaakkola, Regina Barzilay
In this work, we present CATs -- Confident Adaptive Transformers -- in which we simultaneously increase computational efficiency, while guaranteeing a specifiable degree of consistency with the original model with high confidence.
1 code implementation • 17 Feb 2021 • Adam Fisch, Tal Schuster, Tommi Jaakkola, Regina Barzilay
We develop a novel approach to conformal prediction when the target task has limited data available for training.
no code implementations • 9 Nov 2020 • Wengong Jin, Regina Barzilay, Tommi Jaakkola
Drug combinations play an important role in therapeutics due to its better efficacy and reduced toxicity.
1 code implementation • 28 Sep 2020 • Peiyuan Liao, Han Zhao, Keyulu Xu, Tommi Jaakkola, Geoffrey Gordon, Stefanie Jegelka, Ruslan Salakhutdinov
While the advent of Graph Neural Networks (GNNs) has greatly improved node and graph representation learning in many applications, the neighborhood aggregation scheme exposes additional vulnerabilities to adversaries seeking to extract node-level information about sensitive attributes.
1 code implementation • ICLR 2021 • Adam Fisch, Tal Schuster, Tommi Jaakkola, Regina Barzilay
This set is guaranteed to contain a correct answer with high probability, and is well-suited for many open-ended classification tasks.
1 code implementation • 15 Jun 2020 • Karren Yang, Samuel Goldman, Wengong Jin, Alex Lu, Regina Barzilay, Tommi Jaakkola, Caroline Uhler
In this paper, we aim to synthesize cell microscopy images under different molecular interventions, motivated by practical applications to drug development.
2 code implementations • 8 Jun 2020 • Benson Chen, Gary Bécigneul, Octavian-Eugen Ganea, Regina Barzilay, Tommi Jaakkola
Current graph neural network (GNN) architectures naively average or sum node embeddings into an aggregated graph representation -- potentially losing structural or semantic information.
Ranked #1 on
Graph Regression
on Lipophilicity
(using extra training data)
no code implementations • 6 Jun 2020 • Wengong Jin, Regina Barzilay, Tommi Jaakkola
We evaluate our method on multiple applications: molecular property prediction, protein homology and stability prediction and show that RGM significantly outperforms previous state-of-the-art baselines.
no code implementations • 5 May 2020 • Wengong Jin, Regina Barzilay, Tommi Jaakkola
Effective property prediction methods can help accelerate the search for COVID-19 antivirals either through accurate in-silico screens or by effectively guiding on-going at-scale experimental efforts.
no code implementations • ICLR 2020 • Chen-Yu Hsu, Abbas Zeitoun, Guang-He Lee, Dina Katabi, Tommi Jaakkola
We show that this cross-modal prediction task allows us to detect when a particular appliance is used, and the location of the appliance in the home, all in a self-supervised manner, without any labeled data.
1 code implementation • 20 Feb 2020 • Shangyuan Tong, Timur Garipov, Tommi Jaakkola
We provide sufficient conditions for local convergence; characterize the capacity balance that should guide the discriminator and generator choices; and construct examples of minimally sufficient discriminators.
no code implementations • ICML 2020 • Vikas K. Garg, Stefanie Jegelka, Tommi Jaakkola
We address two fundamental questions about graph neural networks (GNNs).
2 code implementations • ICML 2020 • Kevin Yang, Wengong Jin, Kyle Swanson, Regina Barzilay, Tommi Jaakkola
The property predictor is then used as a likelihood model for filtering candidate structures from the generative model.
2 code implementations • ICML 2020 • Wengong Jin, Regina Barzilay, Tommi Jaakkola
Indeed, as we demonstrate, their performance degrades significantly for larger molecules.
1 code implementation • EMNLP 2020 • Tianxiao Shen, Victor Quach, Regina Barzilay, Tommi Jaakkola
We propose Blank Language Model (BLM), a model that generates sequences by dynamically creating and filling in blanks.
2 code implementations • 8 Feb 2020 • Wengong Jin, Regina Barzilay, Tommi Jaakkola
These rationales are identified from molecules as substructures that are likely responsible for each property of interest.
1 code implementation • ICLR Workshop DeepGenStruct 2019 • John Ingraham, Vikas Garg, Regina Barzilay, Tommi Jaakkola
Engineered proteins offer the potential to solve many problems in biomedicine, energy, and materials science, but creating designs that succeed is difficult in practice.
1 code implementation • NeurIPS 2019 • Guy Lorberbom, Tommi Jaakkola, Andreea Gane, Tamir Hazan
Reparameterization of variational auto-encoders with continuous random variables is an effective method for reducing the variance of their gradient estimates.
no code implementations • 25 Sep 2019 • Kevin Yang, Wengong Jin, Kyle Swanson, Regina Barzilay, Tommi Jaakkola
Many challenging prediction problems, from molecular optimization to program synthesis, involve creating complex structured objects as outputs.
no code implementations • 25 Sep 2019 • Tianxiao Shen, Jonas Mueller, Regina Barzilay, Tommi Jaakkola
Neural language models have recently shown impressive gains in unconditional text generation, but controllable generation and manipulation of text remain challenging.
1 code implementation • 11 Jun 2019 • Wengong Jin, Regina Barzilay, Tommi Jaakkola
The problem of accelerating drug discovery relies heavily on automatic tools to optimize precursor molecules to afford them with better biochemical properties.
Ranked #1 on
Drug Discovery
on QED
no code implementations • 29 May 2019 • Vikas K. Garg, Tommi Jaakkola
We introduce a new class of context dependent, incomplete information games to serve as structured prediction models for settings with significant strategic interactions.
2 code implementations • 29 May 2019 • Benson Chen, Regina Barzilay, Tommi Jaakkola
Much of the recent work on learning molecular representations has been based on Graph Convolution Networks (GCN).
3 code implementations • ICML 2020 • Tianxiao Shen, Jonas Mueller, Regina Barzilay, Tommi Jaakkola
We prove that this simple modification guides the latent space geometry of the resulting model by encouraging the encoder to map similar texts to similar latent representations.
no code implementations • NeurIPS 2019 • Vikas K. Garg, Tommi Jaakkola
The transport problem is seeded with prior information about node importance, attributes, and edges in the graph.
no code implementations • ICLR 2019 • Wengong Jin, Kevin Yang, Regina Barzilay, Tommi Jaakkola
We evaluate our model on multiple molecule optimization tasks and show that our model outperforms previous state-of-the-art baselines by a significant margin.
no code implementations • ICLR 2019 • Paresh Malalur, Tommi Jaakkola
An attention mechanism can be used to highlight the area of the image that the model focuses on thus offering a narrow view into the mechanism of classification.
5 code implementations • 2 Apr 2019 • Kevin Yang, Kyle Swanson, Wengong Jin, Connor Coley, Philipp Eiden, Hua Gao, Angel Guzman-Perez, Timothy Hopper, Brian Kelley, Miriam Mathea, Andrew Palmer, Volker Settels, Tommi Jaakkola, Klavs Jensen, Regina Barzilay
In addition, we introduce a graph convolutional model that consistently matches or outperforms models using fixed molecular descriptors as well as previous graph neural architectures on both public and proprietary datasets.
Ranked #2 on
Molecular Property Prediction
on ClinTox
no code implementations • 11 Mar 2019 • Paresh Malalur, Tommi Jaakkola
An attention mechanism can be used to highlight the area of the image that the model focuses on thus offering a narrow view into the mechanism of classification.
5 code implementations • 3 Dec 2018 • Wengong Jin, Kevin Yang, Regina Barzilay, Tommi Jaakkola
We evaluate our model on multiple molecular optimization tasks and show that our model outperforms previous state-of-the-art baselines.
2 code implementations • NeurIPS 2018 • David Alvarez Melis, Tommi Jaakkola
Most recent work on interpretability of complex machine learning models has focused on estimating a-posteriori explanations for previously trained models around specific predictions.
1 code implementation • 24 Jul 2018 • Luke B. Hewitt, Maxwell I. Nye, Andreea Gane, Tommi Jaakkola, Joshua B. Tenenbaum
However, when this generative model is expressed as a powerful neural network such as a PixelCNN, we show that existing learning techniques typically fail to effectively use latent variables.
2 code implementations • ICLR 2019 • Guy Lorberbom, Andreea Gane, Tommi Jaakkola, Tamir Hazan
We demonstrate empirically the effectiveness of the direct loss minimization technique in variational autoencoders with both unstructured and structured discrete latent variables.
10 code implementations • ICML 2018 • Wengong Jin, Regina Barzilay, Tommi Jaakkola
We evaluate our model on multiple tasks ranging from molecular generation to optimization.
Ranked #1 on
Molecular Graph Generation
on InterBioScreen
no code implementations • ICLR 2018 • Luke Hewitt, Andrea Gane, Tommi Jaakkola, Joshua B. Tenenbaum
Hierarchical Bayesian methods have the potential to unify many related tasks (e. g. k-shot classification, conditional, and unconditional generation) by framing each as inference within a single generative model.
no code implementations • ICLR 2018 • Benson Chen, Connor Coley, Regina Barzilay, Tommi Jaakkola
Deep learning algorithms are increasingly used in modeling chemical processes.
no code implementations • NeurIPS 2017 • Vikas Garg, Tommi Jaakkola
Aggregative games provide a rich abstraction to model strategic multi-agent interactions.
1 code implementation • NeurIPS 2017 • Wengong Jin, Connor W. Coley, Regina Barzilay, Tommi Jaakkola
The prediction of organic reaction outcomes is a fundamental problem in computational chemistry.
no code implementations • ICML 2017 • Jonas Mueller, David Gifford, Tommi Jaakkola
Under this model, gradient methods can be used to efficiently optimize the continuous latent factors with respect to inferred outcomes.
1 code implementation • 1 Aug 2017 • Karthik Narasimhan, Regina Barzilay, Tommi Jaakkola
In this paper, we explore the utilization of natural language to drive transfer for reinforcement learning (RL).
12 code implementations • NeurIPS 2017 • Tianxiao Shen, Tao Lei, Regina Barzilay, Tommi Jaakkola
We demonstrate the effectiveness of this cross-alignment method on three tasks: sentiment modification, decipherment of word substitution ciphers, and recovery of word order.
Ranked #6 on
Text Style Transfer
on Yelp Review Dataset (Small)
no code implementations • ICML 2017 • Tao Lei, Wengong Jin, Regina Barzilay, Tommi Jaakkola
The design of neural architectures for structured objects is typically guided by experimental insights rather than a formal process.
1 code implementation • TACL 2017 • Yuan Zhang, Regina Barzilay, Tommi Jaakkola
We introduce a neural method for transfer learning between two (source and target) classification tasks or aspects over the same domain.
no code implementations • NeurIPS 2016 • Vikas Garg, Tommi Jaakkola
Many real phenomena, including behaviors, involve strategic interactions that can be learned from data.
no code implementations • 16 Jun 2016 • Jonas Mueller, David N. Reshef, George Du, Tommi Jaakkola
Assuming the underlying relationship remains invariant under intervention, we develop efficient algorithms to identify the optimal intervention policy from limited data and provide theoretical guarantees for our approach in a Gaussian Process setting.
3 code implementations • EMNLP 2016 • Tao Lei, Regina Barzilay, Tommi Jaakkola
Our approach combines two modular components, generator and encoder, which are trained to operate well together.
no code implementations • 10 Feb 2016 • Tamir Hazan, Francesco Orabona, Anand D. Sarwate, Subhransu Maji, Tommi Jaakkola
This paper shows that the expected value of perturb-max inference with low dimensional perturbations can be used sequentially to generate unbiased samples from the Gibbs distribution.
1 code implementation • NAACL 2016 • Tao Lei, Hrishikesh Joshi, Regina Barzilay, Tommi Jaakkola, Katerina Tymoshenko, Alessandro Moschitti, Lluis Marquez
Question answering forums are rapidly growing in size with no effective automated ability to refer to and reuse answers already available for previous posted questions.
no code implementations • NeurIPS 2015 • Jonas Mueller, Tommi Jaakkola
We introduce principal differences analysis (PDA) for analyzing differences between high-dimensional distributions.
no code implementations • 20 Aug 2015 • Tamir Hazan, Tommi Jaakkola
Contemporary deep neural networks exhibit impressive results on practical problems.
2 code implementations • EMNLP 2015 • Tao Lei, Regina Barzilay, Tommi Jaakkola
Moreover, we extend the n-gram convolution to non-consecutive words to recognize patterns with intervening words.
no code implementations • 5 Aug 2015 • Jean Honorio, Tommi Jaakkola
Thus, using the maximum loss over random structured outputs is a principled way of learning the parameter of structured prediction models.
no code implementations • 25 Jun 2015 • Vikas K. Garg, Cynthia Rudin, Tommi Jaakkola
We present a framework for clustering with cluster-specific feature selection.
1 code implementation • TACL 2015 • Karthik Narasimhan, Regina Barzilay, Tommi Jaakkola
In contrast, we propose a model for unsupervised morphological analysis that integrates orthographic and semantic views of words.
no code implementations • NeurIPS 2014 • Yu Xin, Tommi Jaakkola
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.
no code implementations • NeurIPS 2013 • Tamir Hazan, Subhransu Maji, Joseph Keshet, Tommi Jaakkola
In this work we develop efficient methods for learning random MAP predictors for structured label problems.
no code implementations • 15 Oct 2013 • Francesco Orabona, Tamir Hazan, Anand D. Sarwate, Tommi Jaakkola
Applying the general result to MAP perturbations can yield a more efficient algorithm to approximate sampling from the Gibbs distribution.
no code implementations • NeurIPS 2013 • Tamir Hazan, Subhransu Maji, Tommi Jaakkola
In this paper we describe how MAP inference can be used to sample efficiently from Gibbs distributions.
no code implementations • 25 Aug 2012 • Fahiem Bacchus, Tommi Jaakkola
This is the Proceedings of the Twenty-First Conference on Uncertainty in Artificial Intelligence, which was held in Edinburgh, Scotland July 26 - 29 2005.
no code implementations • 18 Jul 2012 • Jean Honorio, Tommi Jaakkola, Dimitris Samaras
In this paper, we present $\ell_{1, p}$ multi-task structure learning for Gaussian graphical models.