arXiv subjects

Alfredo Braunstein

Publications and source records attributed to Alfredo Braunstein.

44 records · Page 3Linked to original sources

arXiv2011

Efficient LDPC Codes over GF(q) for Lossy Data Compression

In this paper we consider the lossy compression of a binary symmetric source. We present a scheme that provides a low complexity lossy compressor with near optimal empirical performance. The proposed scheme is based on b-reduced ultra-sparse LDPC codes over GF(q). Encoding is performed by the Reinforced Belief Propagation algorithm, a variant of Belief Propagation. The computational complexity at the encoder is O( .n.q.log q), where is the average degree of the check nodes. For our code ensemble, decoding can be performed iteratively following the inverse steps of the leaf removal algorithm. For a sparse parity-check matrix the number of needed operations is O(n).

cs.IT↗

arXiv2011

Statistical physics of optimization under uncertainty

Optimization under uncertainty deals with the problem of optimizing stochastic cost functions given some partial information on their inputs. These problems are extremely difficult to solve and yet pervade all areas of technological and natural sciences. We propose a general approach to solve such large-scale stochastic optimization problems and a Survey Propagation based algorithm that implements it. In the problems we consider some of the parameters are not known at the time of the first optimization, but are extracted later independently of each other from known distributions. As an illustration, we apply our method to the stochastic bipartite matching problem, in the two-stage and multi-stage cases. The efficiency of our approach, which does not rely on sampling techniques, allows us to validate the analytical predictions with large-scale numerical simulations.

cond-mat.stat-mech↗

arXiv2011

Stochastic optimization by message passing

Most optimization problems in applied sciences realistically involve uncertainty in the parameters defining the cost function, of which only statistical information is known beforehand. In a recent work we introduced a message passing algorithm based on the cavity method of statistical physics to solve the two-stage matching problem with independently distributed stochastic parameters. In this paper we provide an in-depth explanation of the general method and caveats, show the details of the derivation and resulting algorithm for the matching problem and apply it to a stochastic version of the independent set problem, which is a computationally hard and relevant problem in communication networks. We compare the results with some greedy algorithms and briefly discuss the extension to more complicated stochastic multi-stage problems.

cond-mat.stat-mech↗

arXiv2011

The stochastic matching problem

The matching problem plays a basic role in combinatorial optimization and in statistical mechanics. In its stochastic variants, optimization decisions have to be taken given only some probabilistic information about the instance. While the deterministic case can be solved in polynomial time, stochastic variants are worst-case intractable. We propose an efficient method to solve stochastic matching problems which combines some features of the survey propagation equations and of the cavity method. We test it on random bipartite graphs, for which we analyze the phase diagram and compare the results with exact bounds. Our approach is shown numerically to be effective on the full range of parameters, and to outperform state-of-the-art methods. Finally we discuss how the method can be generalized to other problems of optimization under uncertainty.

cond-mat.stat-mech↗

arXiv2007

Estimating the size of the solution space of metabolic networks

In this work we propose a novel algorithmic strategy that allows for an efficient characterization of the whole set of stable fluxes compatible with the metabolic constraints. The algorithm, based on the well-known Bethe approximation, allows the computation in polynomial time of the volume of a non full-dimensional convex polytope in high dimensions. The result of our algorithm match closely the prediction of Monte Carlo based estimations of the flux distributions of the Red Blood Cell metabolic network but in incomparably shorter time. We also analyze the statistical properties of the average fluxes of the reactions in the E-Coli metabolic network and finally to test the effect of gene knock-outs on the size of the solution space of the E-Coli central metabolism.

cond-mat.dis-nn↗

arXiv2007

Efficient supervised learning in networks with binary synapses

Recent experimental studies indicate that synaptic changes induced by neuronal activity are discrete jumps between a small number of stable states. Learning in systems with discrete synapses is known to be a computationally hard problem. Here, we study a neurobiologically plausible on-line learning algorithm that derives from Belief Propagation algorithms. We show that it performs remarkably well in a model neuron with binary synapses, and a finite number of `hidden' states per synapse, that has to learn a random classification task. Such system is able to learn a number of associations close to the theoretical limit, in time which is sublinear in system size. This is to our knowledge the first on-line algorithm that is able to achieve efficiently a finite number of patterns learned per binary synapse. Furthermore, we show that performance is optimal for a finite number of hidden states which becomes very small for sparse coding. The algorithm is similar to the standard `perceptron' learning algorithm, with an additional rule for synaptic transitions which occur only if a currently presented pattern is `barely correct'. In this case, the synaptic changes are meta-plastic only (change in hidden states and not in actual synaptic state), stabilizing the synapse in its current state. Finally, we show that a system with two visible states and K hidden states is much more robust to noise than a system with K visible states. We suggest this rule is sufficiently simple to be easily implemented by neurobiological systems or in hardware.

q-bio.NC↗

arXiv2005

Learning by message-passing in networks of discrete synapses

We show that a message-passing process allows to store in binary "material" synapses a number of random patterns which almost saturates the information theoretic bounds. We apply the learning algorithm to networks characterized by a wide range of different connection topologies and of size comparable with that of biological systems (e.g. $n\simeq10^{5}-10^{6}$). The algorithm can be turned into an on-line --fault tolerant-- learning protocol of potential interest in modeling aspects of synaptic plasticity and in building neuromorphic devices.

cond-mat.dis-nn↗

arXiv2004

Source coding by efficient selection of ground states clusters

In this letter, we show how the Survey Propagation algorithm can be generalized to include external forcing messages, and used to address selectively an exponential number of glassy ground states. These capabilities can be used to explore efficiently the space of solutions of random NP-complete constraint satisfaction problems, providing a direct experimental evidence of replica symmetry breaking in large-size instances. Finally, a new lossy data compression protocol is introduced, exploiting as a computational resource the clustered nature of the space of addressable states.

cond-mat.dis-nn↗