If you are looking for an answer to the question What is Artificial Intelligence? and you only have a minute, then here's the definition the Association for the Advancement of Artificial Intelligence offers on its home page: "the scientific understanding of the mechanisms underlying thought and intelligent behavior and their embodiment in machines."
However, if you are fortunate enough to have more than a minute, then please get ready to embark upon an exciting journey exploring AI (but beware, it could last a lifetime) …
As machine learning systems move from computer-science laboratories into the open world, their accountability becomes a high priority problem. Accountability requires deep understanding of system behavior and its failures. Current evaluation methods such as single-score error metrics and confusion matrices provide aggregate views of system performance that hide important shortcomings. Understanding details about failures is important for identifying pathways for refinement, communicating the reliability of systems in different settings, and for specifying appropriate human oversight and engagement. Characterization of failures and shortcomings is particularly complex for systems composed of multiple machine learned components. For such systems, existing evaluation methods have limited expressiveness in describing and explaining the relationship among input content, the internal states of system components, and final output quality. We present Pandora, a set of hybrid human-machine methods and tools for describing and explaining system failures. Pandora leverages both human and system-generated observations to summarize conditions of system malfunction with respect to the input content and system architecture. We share results of a case study with a machine learning pipeline for image captioning that show how detailed performance views can be beneficial for analysis and debugging.
Agents trained in simulation may make errors in the real world due to mismatches between training and execution environments. These mistakes can be dangerous and difficult to discover because the agent cannot predict them a priori. We propose using oracle feedback to learn a predictive model of these blind spots to reduce costly errors in real-world applications. We focus on blind spots in reinforcement learning (RL) that occur due to incomplete state representation: The agent does not have the appropriate features to represent the true state of the world and thus cannot distinguish among numerous states. We formalize the problem of discovering blind spots in RL as a noisy supervised learning problem with class imbalance. We learn models to predict blind spots in unseen regions of the state space by combining techniques for label aggregation, calibration, and supervised learning. The models take into consideration noise emerging from different forms of oracle feedback, including demonstrations and corrections. We evaluate our approach on two domains and show that it achieves higher predictive performance than baseline methods, and that the learned model can be used to selectively query an oracle at execution time to prevent errors. We also empirically analyze the biases of various feedback types and how they influence the discovery of blind spots.
Segal, Avi (Ben-Gurion University of the Negev) | Gal, Kobi (Ben-Gurion University of the Negev) | Kamar, Ece (Microsoft Research, Redmond WA) | Horvitz, Eric (Microsoft Research, Redmond WA) | Miller, Grant (University of Oxford)
Volunteers who help with online crowdsourcing such as citizen science tasks typically make only a few contributions before exiting. We propose a computational approach for increasing users' engagement in such settings that is based on optimizing policies for displaying motivational messages to users. The approach, which we refer to as Trajectory Corrected Intervention (TCI), reasons about the tradeoff between the long-term influence of engagement messages on participants' contributions and the potential risk of disrupting their current work. We combine model-based reinforcement learning with off-line policy evaluation to generate intervention policies, without relying on a fixed representation of the domain. TCI works iteratively to learn the best representation from a set of random intervention trials and to generate candidate intervention policies. It is able to refine selected policies off-line by exploiting the fact that users can only be interrupted once per session.We implemented TCI in the wild with Galaxy Zoo, one of the largest citizen science platforms on the web. We found that TCI was able to outperform the state-of-the-art intervention policy for this domain, and significantly increased the contributions of thousands of users. This work demonstrates the benefit of combining traditional AI planning with off-line policy methods to generate intelligent intervention strategies.
We propose an efficient method to estimate the accuracy of classifiers using only unlabeled data. We consider a setting with multiple classification problems where the target classes may be tied together through logical constraints. For example, a set of classes may be mutually exclusive, meaning that a data instance can belong to at most one of them. The proposed method is based on the intuition that: (i) when classifiers agree, they are more likely to be correct, and (ii) when the classifiers make a prediction that violates the constraints, at least one classifier must be making an error. Experiments on four real-world data sets produce accuracy estimates within a few percent of the true accuracy, using solely unlabeled data. Our models also outperform existing state-of-the-art solutions in both estimating accuracies, and combining multiple classifier outputs. The results emphasize the utility of logical constraints in estimating accuracy, thus validating our intuition.
Vehicle crashes account for over one million fatalities and many more million injuries annually worldwide. Some roads are safer than others, so a driving route optimized for safety may reduce the number of crashes. We have developed a method to estimate the probability of a crash on any road as a function of the traffic volume, road characteristics, and environmental conditions. We trained a regression model to estimate traffic volume and a binary classifier to estimate crash probability on road segments. Modeling a route’s crash probability as a series of Bernoulli probability trials, we show how to use a simple Dijkstra algorithm to compute the safest route between two locations. Compared to the fastest route, the safest route averages about 1.7 times as long in duration and about half as dangerous. We also show how to smoothly trade off safety for time, giving several different route options with different crash probabilities and durations.
Analyses of text corpora over time can reveal trends in beliefs, interest, and sentiment about a topic. We focus on views expressed about artificial intelligence (AI) in the New York Times over a 30-year period. General interest, awareness, and discussion about AI has waxed and waned since the field was founded in 1956. We present a set of measures that captures levels of engagement, measures of pessimism and optimism, the prevalence of specific hopes and concerns, and topics that are linked to discussions about AI over decades. We find that discussion of AI has increased sharply since 2009, and that these discussions have been consistently more optimistic than pessimistic. However, when we examine specific concerns, we find that worries of loss of control of AI, ethical concerns for AI, and the negative impact of AI on work have grown in recent years. We also find that hopes for AI in healthcare and education have increased over time.
Patients in intensive care units (ICU) are acutely ill and have the highest mortality rates for hospitalized patients. Predictive models and planning system could forecast and guide interventions to prevent the hazardous deterioration of patients’ physiologies, thereby giving the opportunity of employing machine learning and inference to assist with the care of ICU patients. We report on the construction of a prediction pipeline that estimates the probability of death by inferring rates of hazard over time, based on patients’ physiological measurements. The inferred model provided the contribution of each variable and information about the influence of sets of observations on the overall risks and expected trajectories of patients.
Predictive models deployed in the real world may assign incorrect labels to instances with high confidence. Such errors or unknown unknowns are rooted in model incompleteness, and typically arise because of the mismatch between training data and the cases encountered at test time. As the models are blind to such errors, input from an oracle is needed to identify these failures. In this paper, we formulate and address the problem of informed discovery of unknown unknowns of any given predictive model where unknown unknowns occur due to systematic biases in the training data.We propose a model-agnostic methodology which uses feedback from an oracle to both identify unknown unknowns and to intelligently guide the discovery. We employ a two-phase approach which first organizes the data into multiple partitions based on the feature similarity of instances and the confidence scores assigned by the predictive model, and then utilizes an explore-exploit strategy for discovering unknown unknowns across these partitions. We demonstrate the efficacy of our framework by varying the underlying causes of unknown unknowns across various applications. To the best of our knowledge, this paper presents the first algorithmic approach to the problem of discovering unknown unknowns of predictive models.
We study the problem of troubleshooting machine learning systems that rely on analytical pipelines of distinct components. Understanding and fixing errors that arise in such integrative systems is difficult as failures can occur at multiple points in the execution workflow. Moreover, errors can propagate, become amplified or be suppressed, making blame assignment difficult. We propose a human-in-the-loop methodology which leverages human intellect for troubleshooting system failures. The approach simulates potential component fixes through human computation tasks and measures the expected improvements in the holistic behavior of the system. The method provides guidance to designers about how they can best improve the system. We demonstrate the effectiveness of the approach on an automated image captioning system that has been pressed into real-world use.
Russell, Stuart (University of California, Berkeley) | Dietterich, Tom (Oregon State University) | Horvitz, Eric (Microsoft) | Selman, Bart (Cornell University) | Rossi, Francesca (University of Padova) | Hassabis, Demis (DeepMind) | Legg, Shane (DeepMind) | Suleyman, Mustafa (DeepMind) | George, Dileep (Vicarious) | Phoenix, Scott (Vicarious)
The adoption of probabilistic and decision-theoretic representations and statistical learning methods has led to a large degree of integration and cross-fertilization among AI, machine learning, statistics, control theory, neuroscience, and other fields. The progress in AI research makes it timely to focus research not only on making AI more capable, but also on maximizing the societal benefit of AI. We recommend expanded research aimed at ensuring that increasingly capable AI systems are robust and beneficial: our AI systems must do what we want them to do. In summary, we believe that research on how to make AI systems robust and beneficial is both important and timely, and that there are concrete research directions that can be pursued today.