Publications

The Director's Office at MIT Lincoln Laboratory (MIT LL) requested a comprehensive study on artificial intelligence (AI) focusing on present applications and future science and technology (S&T) opportunities in the Cyber Security and Information Sciences Division (Division 5). This report elaborates on the main results from the study. Since the AI field is evolving so rapidly, the study scope was to look at the recent past and ongoing developments to lead to a set of findings and recommendations. It was important to begin with a short AI history and a lay-of-the-land on representative developments across the Department of Defense (DoD), intelligence communities (IC), and Homeland Security. These areas are addressed in more detail within the report. A main deliverable from the study was to formulate an end-to-end AI canonical architecture that was suitable for a range of applications. The AI canonical architecture, formulated in the study, serves as the guiding framework for all the sections in this report. Even though the study primarily focused on cyber security and information sciences, the enabling technologies are broadly applicable to many other areas. Therefore, we dedicate a full section on enabling technologies in Section 3. The discussion on enabling technologies helps the reader clarify the distinction among AI, machine learning algorithms, and specific techniques to make an end-to-end AI system viable. In order to understand what is the lay-of-the-land in AI, study participants performed a fairly wide reach within MIT LL and external to the Laboratory (government, commercial companies, defense industrial base, peers, academia, and AI centers). In addition to the study participants (shown in the next section under acknowledgements), we also assembled an internal review team (IRT). The IRT was extremely helpful in providing feedback and in helping with the formulation of the study briefings, as we transitioned from datagathering mode to the study synthesis. The format followed throughout the study was to highlight relevant content that substantiates the study findings, and identify a set of recommendations. An important finding is the significant AI investment by the so-called "big 6" commercial companies. These major commercial companies are Google, Amazon, Facebook, Microsoft, Apple, and IBM. They dominate in the AI ecosystem research and development (R&D) investments within the U.S. According to a recent McKinsey Global Institute report, cumulative R&D investment in AI amounts to about $30 billion per year. This amount is substantially higher than the R&D investment within the DoD, IC, and Homeland Security. Therefore, the DoD will need to be very strategic about investing where needed, while at the same time leveraging the technologies already developed and available from a wide range of commercial applications. As we will discuss in Section 1 as part of the AI history, MIT LL has been instrumental in developing advanced AI capabilities. For example, MIT LL has a long history in the development of human language technologies (HLT) by successfully applying machine learning algorithms to difficult problems in speech recognition, machine translation, and speech understanding. Section 4 elaborates on prior applications of these technologies, as well as newer applications in the context of multi-modalities (e.g., speech, text, images, and video). An end-to-end AI system is very well suited to enhancing the capabilities of human language analysis. Section 5 discusses AI's nascent role in cyber security. There have been cases where AI has already provided important benefits. However, much more research is needed in both the application of AI to cyber security and the associated vulnerability to the so-called adversarial AI. Adversarial AI is an area very critical to the DoD, IC, and Homeland Security, where malicious adversaries can disrupt AI systems and make them untrusted in operational environments. This report concludes with specific recommendations by formulating the way forward for Division 5 and a discussion of S&T challenges and opportunities. The S&T challenges and opportunities are centered on the key elements of the AI canonical architecture to strengthen the AI capabilities across the DoD, IC, and Homeland Security in support of national security.

Machine learning methods in Human Language Technology have reached a stage of maturity where widespread use is both possible and desirable. The MIT Lincoln Laboratory LLTools software suite provides a step towards this goal by providing a set of easily accessible frameworks for incorporating speech, text, and entity resolution components into larger applications. For the speech processing component, the pySLGR (Speaker, Language, Gender Recognition) tool provides signal processing, standard feature analysis, speech utterance embedding, and machine learning modeling methods in Python. The text processing component in LLTools extracts semantically meaningful insights from unstructured data via entity extraction, topic modeling, and document classification. The entity resolution component in LLTools provides approximate string matching, author recognition and graph-based methods for identifying and linking different instances of the same real-world entity. We show through two applications that LLTools can be used to rapidly create and train research prototypes for human language processing.

Patent litigation is an expensive and time-consuming process. To minimize its impact on the participants in the patent lifecycle, automatic determination of litigation potential is a compelling machine learning application. In this paper, we consider preliminary methods for the prediction of a patent being involved in litigation using metadata, content, and graph features. Metadata features are top-level easily-extractable features, i.e., assignee, number of claims, etc. The content feature performs lexical analysis of the claims associated to a patent. Graph features use relational learning to summarize patent references. We apply our methods on US patents using a labeled data set. Prior work has focused on metadata-only features, but we show that both graph and content features have significant predictive capability. Additionally, fusing all features results in improved performance. We also perform a preliminary examination of some of the qualitative factors that may have significant importance in patent litigation.

Automatic extraction of intelligent and useful information from data is one of the main goals in data science. Traditional approaches have focused on learning from structured features, i.e., information in a relational database. However, most of the data encountered in practice are unstructured (i.e., social media posts, forums, emails and web logs); they do not have a predefined schema or format. In this work, we examine unsupervised methods for processing unstructured text data, extracting relevant information, and transforming it into structured information that can then be leveraged in various applications such as graph analysis and matching entities across different platforms. Various efforts have been proposed to develop algorithms for processing unstructured text data. At a top level, text can be either summarized by document level features (i.e., language, topic, genre, etc.) or analyzed at a word or sub-word level. Text analytics can be unsupervised, semi-supervised, or supervised. In this work, we focus on word analysis and unsupervised methods. Unsupervised (or semi-supervised) methods require less human annotation and can easily fulfill the role of automatic analysis. For text analysis, we focus on methods for finding relevant words in the text. Specifically, we look at social media data and attempt to predict hashtags for users' posts. The resulting hashtags can be used for downstream processing such as graph analysis. Automatic hashtag annotation is closely related to automatic tag extraction and keyword extraction. Techniques for hashtags extraction include topic analysis, supervised classifiers, machine translation methods, and collaborative filtering. Methods for keyword extraction include graph-based and topical analysis of text.

A new DARPA program called Data Driven Discovery of Models (D3M) aims to develop automated model discovery systems that can be used by researchers with specific subject matter expertise to create empirical models of real, complex processes. Two major goals of this program are to allow experts to create empirical models without the need for data scientists and to increase the productivity of data scientists via automation. Automated model discovery systems developed will be tested on real-world problems that progressively get harder during the course of the program. Toward the end of the program, problems will be both unsolved and underspecified in terms of data and desired outcomes. The program will emphasize creating and leveraging open source technology and architecture. Our presentation reviews the goals and structure of this program which will begin early in 2017. Although the deadline for submitting proposals has past, we welcome suggestions concerning challenge tasks, evaluations, or new open-source data sets to be included for system development and evaluation that would supplement data currently being curated from many sources.

The automatic determination of emotional state from multimedia content is an inherently challenging problem with a broad range of applications including biomedical diagnostics, multimedia retrieval, and human computer interfaces. The Audio Video Emotion Challenge (AVEC) 2016 provides a well-defined framework for developing and rigorously evaluating innovative approaches for estimating the arousal and valence states of emotion as a function of time. It presents the opportunity for investigating multimodal solutions that include audio, video, and physiological sensor signals. This paper provides an overview of our AVEC Emotion Challenge system, which uses multi-feature learning and fusion across all available modalities. It includes a number of technical contributions, including the development of novel high- and low-level features for modeling emotion in the audio, video, and physiological channels. Low-level features include modeling arousal in audio with minimal prosodic-based descriptors. High-level features are derived from supervised and unsupervised machine learning approaches based on sparse coding and deep learning. Finally, a state space estimation approach is applied for score fusion that demonstrates the importance of exploiting the time-series nature of the arousal and valence states. The resulting system outperforms the baseline systems [10] on the test evaluation set with an achieved Concordant Correlation Coefficient (CCC) for arousal of 0.770 vs 0.702 (baseline) and for valence of 0.687 vs 0.638. Future work will focus on exploiting the time-varying nature of individual channels in the multi-modal framework.

As an unsupervised learning method, sparse coding can discover high-level representations for an input in a large variety of learning problems. Under semi-supervised settings, sparse coding is used to extract features for a supervised task such as classification. While sparse representations learned from unlabeled data independently of the supervised task perform well, we argue that sparse coding should also be built as a holistic learning unit optimizing on the supervised task objectives more explicitly. In this paper, we propose sparse-coded net, a feedforward model that integrates sparse coding and task-driven output layers, and describe training methods in detail. After pretraining a sparse-coded net via semi-supervised learning, we optimize its task-specific performance in a novel backpropagation algorithm that can traverse nonlinear feature pooling operators to update the dictionary. Thus, sparse-coded net can be applied to supervised dictionary learning. We evaluate sparse-coded net with classification problems in sound, image, and text data. The results confirm a significant improvement over semi-supervised learning as well as superior classification performance against deep stacked autoencoder neural network and GMM-SVM pipelines in small to medium-scale settings.

Large unstructured audio data sets have become ubiquitous and present a challenge for organization and search. One logical approach for structuring data is to find common speakers and link occurrences across different recordings. Prior approaches to this problem have focused on basic methodology for the linking task. In this paper, we introduce a novel trainable nonparametric hashing method for indexing large speaker recording data sets. This approach leads to tunable computational complexity methods for speaker linking. We focus on a scalable clustering method based on hashing canopy-clustering. We apply this method to a large corpus of speaker recordings, demonstrate performance tradeoffs, and compare to other hashing methods.

Spoken language recognition requires a series of signal processing steps and learning algorithms to model distinguishing characteristics of different languages. In this paper, we present a sparse discriminative feature learning framework for language recognition. We use sparse coding, an unsupervised method, to compute efficient representations for spectral features from a speech utterance while learning basis vectors for language models. Differentiated from existing approaches in sparse representation classification, we introduce a maximum a posteriori (MAP) adaptation scheme based on online learning that further optimizes the discriminative quality of sparse-coded speech features. We empirically validate the effectiveness of our approach using the NIST LRE 2015 dataset.

The discovery of community structure in networks is a problem of considerable interest in recent years. In online social networks, often times, users are simultaneously involved in multiple social media sites, some of which share common social relationships. It is of great interest to uncover a shared community structure across these networks. However, in reality, users typically identify themselves with different usernames across social media sites. This creates a great difficulty in detecting the community structure. In this paper, we explore several approaches for community detection across online social networks with limited knowledge of username alignment across the networks. We refer to the known alignment of usernames as seeds. We investigate strategies for seed selection and its impact on networks with a different fraction of overlapping vertices. The goal is to study the interplay between network topologies and seed selection strategies, and to understand how it affects the detected community structure. We also propose several measures to assess the performance of community detection and use them to measure the quality of the detected communities in both Twitter-Twitter networks and Twitter-Instagram networks.