Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox

"Recommendation": models, code, and papers

Pragmatic Neural Language Modelling in Machine Translation

Mar 20, 2015
Paul Baltescu, Phil Blunsom

This paper presents an in-depth investigation on integrating neural language models in translation systems. Scaling neural language models is a difficult task, but crucial for real-world applications. This paper evaluates the impact on end-to-end MT quality of both new and existing scaling techniques. We show when explicitly normalising neural models is necessary and what optimisation tricks one should use in such scenarios. We also focus on scalable training algorithms and investigate noise contrastive estimation and diagonal contexts as sources for further speed improvements. We explore the trade-offs between neural models and back-off n-gram models and find that neural models make strong candidates for natural language applications in memory constrained environments, yet still lag behind traditional models in raw translation quality. We conclude with a set of recommendations one should follow to build a scalable neural language model for MT.

* NAACL 2015 

  Access Paper or Ask Questions

Representing and Reasoning about Game Strategies

Jul 21, 2014
Dongmo Zhang, Michael Thielsher

As a contribution to the challenge of building game-playing AI systems, we develop and analyse a formal language for representing and reasoning about strategies. Our logical language builds on the existing general Game Description Language (GDL) and extends it by a standard modality for linear time along with two dual connectives to express preferences when combining strategies. The semantics of the language is provided by a standard state-transition model. As such, problems that require reasoning about games can be solved by the standard methods for reasoning about actions and change. We also endow the language with a specific semantics by which strategy formulas are understood as move recommendations for a player. To illustrate how our formalism supports automated reasoning about strategies, we demonstrate two example methods of implementation\/: first, we formalise the semantic interpretation of our language in conjunction with game rules and strategy rules in the Situation Calculus; second, we show how the reasoning problem can be solved with Answer Set Programming.

  Access Paper or Ask Questions

Online Structured Prediction via Coactive Learning

Jun 27, 2012
Pannaga Shivaswamy, Thorsten Joachims

We propose Coactive Learning as a model of interaction between a learning system and a human user, where both have the common goal of providing results of maximum utility to the user. At each step, the system (e.g. search engine) receives a context (e.g. query) and predicts an object (e.g. ranking). The user responds by correcting the system if necessary, providing a slightly improved -- but not necessarily optimal -- object as feedback. We argue that such feedback can often be inferred from observable user behavior, for example, from clicks in web-search. Evaluating predictions by their cardinal utility to the user, we propose efficient learning algorithms that have ${\cal O}(\frac{1}{\sqrt{T}})$ average regret, even though the learning algorithm never observes cardinal utility values as in conventional online learning. We demonstrate the applicability of our model and learning algorithms on a movie recommendation task, as well as ranking for web-search.

  Access Paper or Ask Questions

Mean Embeddings with Test-Time Data Augmentation for Ensembling of Representations

Jul 14, 2021
Arsenii Ashukha, Andrei Atanov, Dmitry Vetrov

Averaging predictions over a set of models -- an ensemble -- is widely used to improve predictive performance and uncertainty estimation of deep learning models. At the same time, many machine learning systems, such as search, matching, and recommendation systems, heavily rely on embeddings. Unfortunately, due to misalignment of features of independently trained models, embeddings, cannot be improved with a naive deep ensemble like approach. In this work, we look at the ensembling of representations and propose mean embeddings with test-time augmentation (MeTTA) simple yet well-performing recipe for ensembling representations. Empirically we demonstrate that MeTTA significantly boosts the quality of linear evaluation on ImageNet for both supervised and self-supervised models. Even more exciting, we draw connections between MeTTA, image retrieval, and transformation invariant models. We believe that spreading the success of ensembles to inference higher-quality representations is the important step that will open many new applications of ensembling.

  Access Paper or Ask Questions

Improving Social Welfare While Preserving Autonomy via a Pareto Mediator

Jun 07, 2021
Stephen McAleer, John Lanier, Michael Dennis, Pierre Baldi, Roy Fox

Machine learning algorithms often make decisions on behalf of agents with varied and sometimes conflicting interests. In domains where agents can choose to take their own action or delegate their action to a central mediator, an open question is how mediators should take actions on behalf of delegating agents. The main existing approach uses delegating agents to punish non-delegating agents in an attempt to get all agents to delegate, which tends to be costly for all. We introduce a Pareto Mediator which aims to improve outcomes for delegating agents without making any of them worse off. Our experiments in random normal form games, a restaurant recommendation game, and a reinforcement learning sequential social dilemma show that the Pareto Mediator greatly increases social welfare. Also, even when the Pareto Mediator is based on an incorrect model of agent utility, performance gracefully degrades to the pre-intervention level, due to the individual autonomy preserved by the voluntary mediator.

  Access Paper or Ask Questions

On the Ethical Limits of Natural Language Processing on Legal Text

May 25, 2021
Dimitrios Tsarapatsanis, Nikolaos Aletras

Natural language processing (NLP) methods for analyzing legal text offer legal scholars and practitioners a range of tools allowing to empirically analyze law on a large scale. However, researchers seem to struggle when it comes to identifying ethical limits to using NLP systems for acquiring genuine insights both about the law and the systems' predictive capacity. In this paper we set out a number of ways in which to think systematically about such issues. We place emphasis on three crucial normative parameters which have, to the best of our knowledge, been underestimated by current debates: (a) the importance of academic freedom, (b) the existence of a wide diversity of legal and ethical norms domestically but even more so internationally and (c) the threat of moralism in research related to computational law. For each of these three parameters we provide specific recommendations for the legal NLP community. Our discussion is structured around the study of a real-life scenario that has prompted recent debate in the legal NLP research community.

* Accepted at ACL Findings 2021 

  Access Paper or Ask Questions

Emergence of Structural Bias in Differential Evolution

May 10, 2021
Bas van Stein, Fabio Caraffini, Anna V. Kononova

Heuristic optimisation algorithms are in high demand due to the overwhelming amount of complex optimisation problems that need to be solved. The complexity of these problems is well beyond the boundaries of applicability of exact optimisation algorithms and therefore require modern heuristics to find feasible solutions quickly. These heuristics and their effects are almost always evaluated and explained by particular problem instances. In previous works, it has been shown that many such algorithms show structural bias, by either being attracted to a certain region of the search space or by consistently avoiding regions of the search space, on a special test function designed to ensure uniform 'exploration' of the domain. In this paper, we analyse the emergence of such structural bias for Differential Evolution (DE) configurations and, specifically, the effect of different mutation, crossover and correction strategies. We also analyse the emergence of the structural bias during the run-time of each algorithm. We conclude with recommendations of which configurations should be avoided in order to run DE unbiased.

  Access Paper or Ask Questions

Fairness of Exposure in Stochastic Bandits

Mar 03, 2021
Lequn Wang, Yiwei Bai, Wen Sun, Thorsten Joachims

Contextual bandit algorithms have become widely used for recommendation in online systems (e.g. marketplaces, music streaming, news), where they now wield substantial influence on which items get exposed to the users. This raises questions of fairness to the items -- and to the sellers, artists, and writers that benefit from this exposure. We argue that the conventional bandit formulation can lead to an undesirable and unfair winner-takes-all allocation of exposure. To remedy this problem, we propose a new bandit objective that guarantees merit-based fairness of exposure to the items while optimizing utility to the users. We formulate fairness regret and reward regret in this setting, and present algorithms for both stochastic multi-armed bandits and stochastic linear bandits. We prove that the algorithms achieve sub-linear fairness regret and reward regret. Beyond the theoretical analysis, we also provide empirical evidence that these algorithms can fairly allocate exposure to different arms effectively.

  Access Paper or Ask Questions

Enhancing Reading Strategies by Exploring A Theme-based Approach to Literature Surveys

Feb 10, 2021
Tanya Howden, Pierre Le Bras, Thomas S. Methven, Stefano Padilla, Mike J. Chantler

Searching large digital repositories can be extremely frustrating, as common list-based formats encourage users to adopt a convenience-sampling approach that favours chance discovery and random search, over meaningful exploration. We have designed a methodology that allows users to visually and thematically explore corpora, while developing personalised holistic reading strategies. We describe the results of a three-phase qualitative study, in which experienced researchers used our interactive visualisation approach to analyse a set of publications and select relevant themes and papers. Using in-depth semi-structured interviews and stimulated recall, we found that users: (i) selected papers that they otherwise would not have read, (ii) developed a more coherent reading strategy, and (iii) understood the thematic structure and relationships between papers more effectively. Finally, we make six design recommendations to enhance current digital repositories that we have shown encourage users to adopt a more holistic and thematic research approach.

  Access Paper or Ask Questions