153 Hits in 4.0 sec

An Improved Complex Sphere Decoder for V-BLAST Systems

D. Pham, K.R. Pattipati, P.K. Willett, J. Luo
2004 IEEE Signal Processing Letters  
A new complex sphere decoding algorithm is presented for signal detection in V-BLAST systems, which has a computational cost that is significantly lower than that of the original complex sphere decoder  ...  Simulation results for a 64-QAM system with 23 transmit and 23 receive antennas at an SNR per bit of 24dB show that the new sphere decoding algorithm obtains the ML solution with an average cost that is  ...  (n T = n R = 23, 64-QAM, SN R per bit = 26dB, 500 Monte Carlo Runs) Example 3: Average computational cost vs. SNR per bit. (n T = n R = 30, 16-QAM, 500 Monte Carlo Runs)  ... 
doi:10.1109/lsp.2004.833522 fatcat:hjrkoajtxrhl3bedqa6scjncki

Bounded Rational Decision-Making with Adaptive Neural Network Priors [chapter]

Heinke Hihn, Sebastian Gottwald, Daniel A. Braun
2018 Lecture Notes in Computer Science  
Between prior and posterior lies an anytime deliberation process that can be instantiated by sample-based evaluations of the utility function through Markov Chain Monte Carlo (MCMC) optimization.  ...  In this work we investigate generative neural networks as priors that are optimized concurrently with anytime sample-based decision-making processes such as MCMC.  ...  networks and policy networks with Monte Carlo Tree Search, leading to a major breakthrough in reinforcement learning.  ... 
doi:10.1007/978-3-319-99978-4_17 fatcat:qzj6axnqsjby3hipjqjapqcrfq

Path Planning in Dynamic Environments using Generative RNNs and Monte Carlo Tree Search [article]

Stuart Eiffert, He Kong, Navid Pirmarzdashti, Salah Sukkarieh
2020 arXiv   pre-print
To overcome this limitation, this paper proposes an integrated path planning framework using generative Recurrent Neural Networks within a Monte Carlo Tree Search (MCTS).  ...  In this paper we propose an integrated path planning framework using generative Recurrent Neural Networks (RNNs) and Monte Carlo Tree Search (MCTS).  ...  The search tree consists of nodes representing each state, and edges representing actions.  ... 
arXiv:2001.11597v1 fatcat:jx7cs6som5do7drswapangjphy

Integrating Acting, Planning and Learning in Hierarchical Operational Models [article]

Sunandita Patra, James Mason, Amit Kumar, Malik Ghallab, Paolo Traverso, Dana Nau
2020 arXiv   pre-print
Our planning procedure, UPOM, does a UCT-like search in the space of operational models in order to find a near-optimal method to use for the task and context at hand.  ...  This work has been supported in part by NRL grant N00173191G001.  ...  It is a UCT-like (Kocsis and Szepesvári 2006) Monte Carlo tree search procedure over the space of refinement trees for τ (see Figure 1 ).  ... 
arXiv:2003.03932v1 fatcat:y537bk7ihvfjhewry3oittxemq

Accelerating Cooperative Planning for Automated Vehicles with Learned Heuristics and Monte Carlo Tree Search [article]

Karl Kurzer, Marcus Fechner, J. Marius Zöllner
2020 arXiv   pre-print
However, the search space for cooperative plans is so large that most of the computational budget is spent on exploring the search space in unpromising regions that are far away from the solution.  ...  To accelerate the planning process, we combined learned heuristics with a cooperative planning method to guide the search towards regions with promising actions, yielding better solutions at lower computational  ...  Tree Search; During the expansion phase (i.e. the exploration of the actions space), the current state of the Monte Carlo Tree Search (MCTS) is being transformed into a feature vector f , consisting of  ... 
arXiv:2002.00497v2 fatcat:wv67jddfjndund57m3myvjmbym

Deliberative Acting, Online Planning and Learning with Hierarchical Operational Models [article]

Sunandita Patra, James Mason, Malik Ghallab, Dana Nau, Paolo Traverso
2021 arXiv   pre-print
The anytime planner uses a UCT-like Monte Carlo Tree Search procedure, called UPOM, (UCT Procedure for Operational Models), whose rollouts are simulations of the actor's operational models.  ...  The acting component, called Reactive Acting Engine (RAE), is inspired by the well-known PRS system.  ...  This work has been supported in part by NRL grant N00173191G001, ONR grant N000142012257, and AFOSR grant 1010GWA357.  ... 
arXiv:2010.01909v2 fatcat:x5jt7jnptzczfdgbpey3h2qt4m

Survey on Models and Techniques for Root-Cause Analysis [article]

Marc Solé, Victor Muntés-Mulero, Annie Ibrahim Rana, Giovani Estrada
2017 arXiv   pre-print
Dempster-Shafer Theory MPE [145] ✓ ✓ ✓ ✓ ✓ ✓ ✗ O(2 min(m, Θ ) ) Monte-Carlo [146] ✓ ✗ ✓ ✓ ✓ ✓ ✗ O(m Θ ) Fault Tree Tree search Fault tree top-down [147] ✓ ✓ ✗ ✗ ✗ ✗ ✗ O(e + f ) Decision trees Explanation  ...  Search-based [176] ✓ ✗ ✓ ✗ ✓ ✓ ✗ O(n exp(w)) Loopy Belief propagation [177] ✓ ✗ ✓ ✗ ✓ ✓ ✗ O(n exp(w)) Stochastic Sampling [178]-[182] ✓ ✗ ✓ ✗ ✓ ✓ ✗ anytime Markov Chain Monte Carlo [183] ✓ ✗ ✓ ✗ ✓ ✓ ✗  ... 
arXiv:1701.08546v2 fatcat:wqv5vl3ovbe4bjmq32gnni7gg4

Searching for More Efficient Dynamic Programs [article]

Tim Vieira and Ryan Cotterell and Jason Eisner
2021 arXiv   pre-print
We show that in practice, automated search -- like the mental search performed by human programmers -- can find substantial improvements to the initial program.  ...  For instance, a probabilistic parser may marginalize over exponentially many trees to make predictions. Algorithms for such problems often employ dynamic programming and are not always unique.  ...  In this section, we provide two effective search algorithms for approaching this goal: beam search and Monte Carlo tree search.  ... 
arXiv:2109.06966v1 fatcat:2tvbbntgnrgxlaixayqpjjbosa

Adaptive Visual Information Gathering for Autonomous Exploration of Underwater Environments

Eric Guerrero, Francisco Bonin-Font, Gabriel Oliver
2021 IEEE Access  
We solve the information path planning (IPP) problem by means of a novel depth-first (DF) version of the Monte Carlo tree search (MCTS).  ...  The DF-MCTS method has been designed to explore the state-space in a depth-first fashion, provide solution paths of a given length in an anytime manner, and reward smooth paths for field realization with  ...  Carlo tree search (MCTS).  ... 
doi:10.1109/access.2021.3117343 fatcat:5a5422vcgnc2xgabrk4qapx72u

SOLO: Search Online, Learn Offline for Combinatorial Optimization Problems [article]

Joel Oren, Chana Ross, Maksym Lefarov, Felix Richter, Ayal Taitler, Zohar Feldman, Christian Daniel, Dotan Di Castro
2021 arXiv   pre-print
We mitigate these drawbacks by employing our graph-convolutional policies as non-optimal heuristics in a compatible search algorithm, Monte Carlo Tree Search, to significantly improve overall performance  ...  a modified Monte-Carlo Tree Search (MCTS).  ...  A popular approach for global search is Monte-Carlo Tree Search (MCTS).  ... 
arXiv:2104.01646v3 fatcat:mzprf47ajjg2bkxeesfwfor6ym

Planning in Dynamic Environments with Conditional Autoregressive Models [article]

Johanna Hansen, Kyle Kastner, Aaron Courville, Gregory Dudek
2018 arXiv   pre-print
We test our forward-model with a powerful anytime planning method, Monte-Carlo Tree Search (MCTS) (Kocsis & Szepesvári, 2006) .  ...  MCTS works by rolling out many sequences of actions possible future scenarios to acquire an approximate (Monte Carlo) estimate of the value of taking a specific action from a particular state.  ... 
arXiv:1811.10097v1 fatcat:o6v3mqksvze4xk7covat5of654


Rina Dechter, Irina Rish
2003 Journal of the ACM  
The idea is to bound the dimensionality of dependencies created by inference algorithms.  ...  results demonstrating successful performance of the proposed approximation scheme for the MPE task, both on randomly generated problems and on realistic domains such as medical diagnosis and probabilistic decoding  ...  Finally, there is a large family of approximation techniques somewhat orthogonal to local propagation, namely, sampling techniques (Markov-Chain Monte-Carlo, or MCMC mehtods) often applied to approximate  ... 
doi:10.1145/636865.636866 fatcat:dciqejiqkjekhgn5eplye7aia4

Development of a chemiluminescence-based quantitative lateral flow immunoassay for on-field detection of 2,4,6-trinitrotoluene

Mara Mirasoli, Angela Buragina, Luisa Stella Dolci, Massimo Guardigli, Patrizia Simoni, Angel Montoya, Elisabetta Maiolini, Stefano Girotti, Aldo Roda
2012 Analytica Chimica Acta  
Motivated by this, there is a continuous search for computationally efficient optimal or suboptimal detectors.  ...  The effect of the channel matrix condition number in data detection is exploited in order to achieve a decoding complexity lower than the one of already proposed algorithms with similar performance.  ...  This work was supported by FEDER-MEC under grant TEC2006-14204-C02-01.  ... 
doi:10.1016/j.aca.2012.01.036 pmid:22405316 fatcat:kl63gc7pxnbulnbbocj2oqjjo4

GeoSeq2Seq: Information Geometric Sequence-to-Sequence Networks [article]

Alessandro Bay, Biswa Sengupta
2018 arXiv   pre-print
representation of the latent embedding supersedes the non-probabilistic embedding by 10-15\%.  ...  Specifically, the latent embedding offered by a recurrent network is encoded as a Fisher kernel of a parametric Gaussian Mixture Model, a formalism common in computer vision.  ...  In Bayesian statistics, it has found utility in terms of Riemannian Markov Chain Monte Carlo (MCMC) methods (Girolami & Calderhead, 2011) while for computer vision it has resulted in the Fisher kernel  ... 
arXiv:1710.09363v2 fatcat:lkneiqlalzdfrhi4xpad4wltku

Is Policy Learning Overrated?: Width-Based Planning and Active Learning for Atari [article]

Benjamin Ayton, Masataro Asai
2022 arXiv   pre-print
Experimental results in 55 Atari games demonstrate that it outperforms Rollout-IW by 42-to-11 and VAE-IW by 32-to-20.  ...  computed feature vectors for each screen using a hand designed feature set or a variational autoencoder trained on game screens (VAE-IW), and prune screens that do not have novel features during the search  ...  , and has been demonstrated to outperform Monte Carlo tree search based on UCT (Kocsis and Szepesvári 2006) .  ... 
arXiv:2109.15310v2 fatcat:qzcsrwyzg5hghfejot2gzfxeg4
« Previous Showing results 1 — 15 out of 153 results