Robert Mateescu

AND/OR search spaces for graphical models

The paper introduces an AND/OR search space perspective for graphical models that include probabilistic networks (directed or undirected) and constraint networks. In contrast to the traditional (OR) search space view, the AND/OR search tree displays some of the independencies present in the graphical model explicitly and may sometimes reduce the search space exponentially. Indeed, most algorithmic advances in search-based constraint processing and probabilistic inference can be viewed as searching an AND/OR search tree or graph. Familiar parameters such as the depth of a spanning tree, treewidth and pathwidth are shown to play a key role in characterizing the effect of AND/OR search graphs vs. the traditional OR search graphs. We compare memory intensive AND/OR graph search with inference methods, and place various existing algorithms within the AND/OR search space.

Rank modulation for flash memories

We explore a novel data representation scheme for multi-level flash memory cells, in which a set of n cells stores information in the permutation induced by the different charge levels of the individual cells. The only allowed charge-placement mechanism is a dasiapush-to-the-toppsila operation which takes a single cell of the set and makes it the top-charged cell. The resulting scheme eliminates the need for discrete cell levels, as well as overshoot errors, when programming cells. We present unrestricted Gray codes spanning all possible n-cell states and using only dasiapush-to-the-toppsila operations, and also construct balanced Gray codes. We also investigate optimal rewriting schemes for translating arbitrary input alphabet into n-cell states which minimize the number of programming operations.

Mixed deterministic and probabilistic networks

The paper introduces mixed networks, a new graphical model framework for expressing and reasoning with probabilistic and deterministic information. The motivation to develop mixed networks stems from the desire to fully exploit the deterministic information (constraints) that is often present in graphical models. Several concepts and algorithms specific to belief networks and constraint networks are combined, achieving computational efficiency, semantic coherence and user-interface convenience. We define the semantics and graphical representation of mixed networks, and discuss the two main types of algorithms for processing them: inference-based and search-based. A preliminary experimental evaluation shows the benefits of the new model.

Join-graph propagation algorithms

The paper investigates parameterized approximate message-passing schemes that are based on bounded inference and are inspired by Pearls belief propagation algorithm (BP). We start with the bounded inference mini-clustering algorithm and then move to the iterative scheme called Iterative Join-Graph Propagation (IJGP), that combines both iteration and bounded inference. Algorithm IJGP belongs to the class of Generalized Belief Propagation algorithms, a framework that allowed connections with approximate algorithms from statistical physics and is shown empirically to surpass the performance of mini-clustering and belief propagation, as well as a number of other state-of-the-art algorithms on several classes of networks. We also provide insight into the accuracy of iterative BP and IJGP by relating these algorithms to well known classes of constraint propagation schemes.

AND/OR multi-valued decision diagrams (AOMDDs) for graphical models

Inspired by the recently introduced framework of AND/OR search spaces for graphical models, we propose to augment Multi-Valued Decision Diagrams (MDD) with AND nodes, in order to capture function decomposition structure and to extend these compiled data structures to general weighted graphical models (e.g., probabilistic models). We present the AND/OR Multi-Valued Decision Diagram (AOMDD) which compiles a graphical model into a canonical form that supports polynomial (e.g., solution counting, belief updating) or constant time (e.g. equivalence of graphical models) queries. We provide two algorithms for compiling the AOMDD of a graphical model. The first is search-based, and works by applying reduction rules to the trace of the memory intensive AND/OR search algorithm. The second is inference-based and uses a Bucket Elimination schedule to combine the AOMDDs of the input functions via the the APPLY operator. For both algorithms, the compilation time and the size of the AOMDD are, in the worst case, exponential in the treewidth of the graphical model, rather than pathwidth as is known for ordered binary decision diagrams (OBDDs). We introduce the concept of semantic treewidth, which helps explain why the size of a decision diagram is often much smaller than the worst case bound. We provide an experimental evaluation that demonstrates the potential of AOMDDs.

Repair-optimal MDS array codes over GF(2)

Maximum-distance separable (MDS) array codes with high rate and an optimal repair property were introduced recently. These codes could be applied in distributed storage systems, where they minimize the communication and disk access required for the recovery of failed nodes. However, the encoding and decoding algorithms of the proposed codes use arithmetic over finite fields of order greater than 2, which could result in a complex implementation. In this work, we present a construction of 2-parity MDS array codes, that allow for optimal repair of a failed information node using XOR operations only. The reduction of the field order is achieved by allowing more parity bits to be updated when a single information bit is being changed by the user.

Coding scheme for 3D vertical flash memory

Recently introduced 3D vertical flash memory is expected to be a disruptive technology since it overcomes scaling challenges of conventional 2D planar flash memory by stacking up cells in the vertical direction. However, 3D vertical flash memory suffers from a new problem known as fast detrapping, which is a rapid charge loss problem. In this paper, we propose a scheme to compensate the effect of fast detrapping by intentional inter-cell interference (ICI). In order to properly control the intentional ICI, our scheme relies on a coding technique that incorporates the side information of fast detrapping during the encoding stage. This technique is closely connected to the well-known problem of coding in a memory with defective cells. Numerical results show that the proposed scheme can effectively address the problem of fast detrapping.

Storage coding for wear leveling in flash memories

Flash memory is a nonvolatile computer memory comprised of blocks of cells, wherein each cell is implemented as either NAND or NOR floating gate. NAND flash is currently the most widely used type of flash memory. In a NAND flash memory, every block of cells consists of numerous pages; rewriting even a single page requires the whole block to be erased and reprogrammed. Block erasures determine both the longevity and the efficiency of a flash memory. Therefore, when data in a NAND flash memory are reorganized, minimizing the total number of block erasures required to achieve the desired data movement is an important goal. This leads to the flash data movement problem studied in this paper. We show that coding can significantly reduce the number of block erasures required for data movement, and present several optimal or nearly optimal data-movement algorithms based upon ideas from coding theory and combinatorics. In particular, we show that the sorting-based (noncoding) schemes require O(n log n) erasures to move data among n blocks, whereas coding-based schemes require only O(n) erasures. Furthermore, coding-based schemes use only one auxiliary block, which is the best possible and achieve a good balance between the number of erasures in each of the n+1 blocks.

Data movement in flash memories

NAND flash memories are the most widely used non-volatile memories, and data movement is common in flash storage systems. We study data movement solutions that minimize the number of block erasures, which are very important for the efficiency and longevity of flash memories. To move data among n blocks with the help of ¿ auxiliary blocks, where every block contains m pages, we present algorithms that use ¿(n · min{m, log¿ n}) erasures without the tool of coding. We prove this is almost the best possible for non-coding solutions by presenting a nearly matching lower bound. Optimal data movement can be achieved using coding, where only ¿(n) erasures are needed. We present a coding-based algorithm, which has very low coding complexity, for optimal data movement. We further show the NP hardness of both coding-based and non-coding schemes when the objective is to optimize data movement on a per instance basis.

Locally rewritable codes for resistive memories

We propose locally rewritable codes (LWC) for resistive memories inspired by locally repairable codes (LRC) for distributed storage systems. Small values of repair locality of LRC enable fast repair of a single failed node since the lost data in the failed node can be recovered by accessing only a small fraction of other nodes. By using rewriting locality, LWC can improve endurance and power consumption which are major challenges for resistive memories. We point out the duality between LRC and LWC, which indicates that existing construction methods of LRC can be applied to construct LWC.