Results 1  10
of
26
Consistency of Feature Markov Processes
, 2010
"... We are studying long term sequence prediction (forecasting). We approach this by investigating criteria for choosing a compact useful state representation. The state is supposed to summarize useful information from the history. We want a method that is asymptotically consistent in the sense it will ..."
Abstract

Cited by 9 (7 self)
 Add to MetaCart
We are studying long term sequence prediction (forecasting). We approach this by investigating criteria for choosing a compact useful state representation. The state is supposed to summarize useful information from the history. We want a method that is asymptotically consistent in the sense it will provably eventually only choose between alternatives that satisfy an optimality property related to the used criterion. We extend our work to the case where there is side information that one can take advantage of and, furthermore, we briefly discuss the active setting where an agent takes actions to achieve desirable outcomes.
Probabilistic parsing
 New Developments in Formal Languages and Applications, Studies in Computational Intelligence
, 2008
"... A paper in a previous volume [1] explained parsing, which is the process of determining the parses of an input string according to a formal grammar. Also discussed was tabular parsing, which solves the task of parsing in polynomial time by a form of dynamic programming. In passing, we also mentioned ..."
Abstract

Cited by 5 (0 self)
 Add to MetaCart
(Show Context)
A paper in a previous volume [1] explained parsing, which is the process of determining the parses of an input string according to a formal grammar. Also discussed was tabular parsing, which solves the task of parsing in polynomial time by a form of dynamic programming. In passing, we also mentioned that
THE THEORY OF TRACKABILITY AND ROBUSTNESS FOR PROCESS DETECTION
, 2006
"... Many applications of current interests involve detecting instances of processes from databases or streams of sensor reports. Detecting processes relies on identifying evidences for the existence of such processes from usually noisy and incomplete observable events through statistical inferences. The ..."
Abstract

Cited by 4 (0 self)
 Add to MetaCart
Many applications of current interests involve detecting instances of processes from databases or streams of sensor reports. Detecting processes relies on identifying evidences for the existence of such processes from usually noisy and incomplete observable events through statistical inferences. The performance of inferences can vary dramatically, depending on the complexity of processes ’ behavioral patterns, sensor resolution and sampling rate, SNR, location and coverage, and so on. Stochastic models are mathematical representations of all these factors. In this dissertation, we intend to answer the following questions: Performance – How accurate are the inference results given the model? Trackability – What are the boundaries of the performance of inferences? Robustness – How sensitive is the performance of inferences to perturbations on input data or model parameters? Methodology – How can we improve the trackability and robustness of process detection? From the information theoretic point of view, we address the reason of errors in detection to the losses of source information during the sensing stage, measured as entropy in the Shannon sense. We propose a series of entropic measures of the trackability and robustness for a popular modeling technique – hidden Markov models (HMM). Our major contributions include: the theory of trackability; structural analysis of trackability for HMMs through its nonparametric counterpart – DFA/NFAs; an effective visualization method for analyzing the trackability for
M.: Shingled graph disassembly: Finding the undecideable path
, 2013
"... A probabilistic finite state machine approach to statically disassembling x86 machine language programs is presented and evaluated. Static disassembly is a crucial prerequisite for software reverse engineering, and has many applications in computer security and binary analysis. The general problem i ..."
Abstract

Cited by 3 (2 self)
 Add to MetaCart
(Show Context)
A probabilistic finite state machine approach to statically disassembling x86 machine language programs is presented and evaluated. Static disassembly is a crucial prerequisite for software reverse engineering, and has many applications in computer security and binary analysis. The general problem is provably undecidable because of the heavy use of unaligned instruction encodings and dynamically computed control flows in the x86 architecture. Limited work in machine learning and data mining has been undertaken on this subject. This paper shows that semantic meanings of opcode sequences can be leveraged to infer similarities between groups of opcode and operand sequences. This empowers a probabilistic finite state machine to learn statistically significant opcode and operand sequences in a training corpus of disassemblies. The similarities demonstrate the statistical significance of opcodes and operands in a surrounding context, facilitating more accurate disassembly of new binaries. Empirical results demonstrate that the algorithm is more efficient and effective than comparable approaches used by stateoftheart disassembly tools. 1
SPEDE: Probabilistic edit distance metrics for . . .
, 2012
"... This paper describes Stanford University’s submission ... probabilistic edit distance as predictions of translation quality. We learn weighted edit distance in a probabilistic finite state machine (pFSM) model, where state transitions correspond to edit operations. While standard edit distance model ..."
Abstract

Cited by 2 (1 self)
 Add to MetaCart
This paper describes Stanford University’s submission ... probabilistic edit distance as predictions of translation quality. We learn weighted edit distance in a probabilistic finite state machine (pFSM) model, where state transitions correspond to edit operations. While standard edit distance models cannot capture longdistance word swapping or cross alignments, we rectify these shortcomings using a novel pushdown automaton extension of the pFSM model. Our models are trained in a regression framework, and can easily incorporate a rich set of linguistic features. Evaluated on two different prediction tasks across a diverse set of datasets, our methods achieve stateoftheart correlation with human judgments.
Improving SpeechtoSpeech Translation Using Word Posterior Probabilities
"... Nowadays, speech translation is a research problem in machine translation. The problem arises as to how to combine speech recognition and machine translation in a suitable way. Some authors have shown that the speech translation can be improved by using word lattices as input of the translation syst ..."
Abstract

Cited by 1 (0 self)
 Add to MetaCart
(Show Context)
Nowadays, speech translation is a research problem in machine translation. The problem arises as to how to combine speech recognition and machine translation in a suitable way. Some authors have shown that the speech translation can be improved by using word lattices as input of the translation system. The acoustic recognition scores from the word lattice are used for improving the translation quality. However, word lattices do not consider word cooccurrences between different hypothesis and those probabilities are not real probabilities but merely Viterbi approximations. In this work, we propose an improved word lattice representation for using posterior probabilities instead of acoustic scores. We present preliminary results of this approach compared against other common approaches on two different corpora. Although the results are not strongly conclusive, they show that this approach is worth exploring more deeply. 1
Using Word Posterior Probabilities in Lattice Translation
"... In this paper we describe the statistical machine translation system developed at ITI/UPV, which aims especially at speech recognition and statistical machine translation integration, for the evaluation campaign of the International Workshop on Spoken Language Translation (2007). The system we have ..."
Abstract

Cited by 1 (0 self)
 Add to MetaCart
(Show Context)
In this paper we describe the statistical machine translation system developed at ITI/UPV, which aims especially at speech recognition and statistical machine translation integration, for the evaluation campaign of the International Workshop on Spoken Language Translation (2007). The system we have developed takes advantage of an improved word lattice representation that uses word posterior probabilities. These word posterior probabilities are then added as a feature to a loglineal model. This model includes a stochastic finitestate transducer which allows an easy lattice integration. Furthermore, it provides a statistical phrasebased reordering model that is able to perform local reorderings of the output. We have tested this model on the ItalianEnglish corpus, for clean text, 1best ASR and lattice ASR inputs. The results and conclusions of such experiments are reported at the end of this paper. 1.
SPEDE: Probabilistic edit distance . . .
"... This paper describes Stanford University’s submission to the Shared Evaluation Task of WMT... probabilistic edit distance as predictions of translation quality. We learn weighted edit distance in a probabilistic finite state machine (pFSM) model, where state transitions correspond to edit operation ..."
Abstract
 Add to MetaCart
This paper describes Stanford University’s submission to the Shared Evaluation Task of WMT... probabilistic edit distance as predictions of translation quality. We learn weighted edit distance in a probabilistic finite state machine (pFSM) model, where state transitions correspond to edit operations. While standard edit distance models cannot capture longdistance word swapping or cross alignments, we rectify these shortcomings using a novel pushdown automaton extension of the pFSM model. Our models are trained in a regression framework, and can easily incorporate a rich set of linguistic features. Evaluated on two different prediction tasks across a diverse set of datasets, our methods achieve stateoftheart correlation with human judgments.