absorb.md

Peter Shor

Chronological feed of everything captured from Peter Shor.

Quantum State Restoration Enables Subsystem Cloning and Tomography from Single Copy with Verifier

Given a single copy of an n-qubit state |ψ⟩ and a verification procedure (e.g., measuring |ψ⟩⟨ψ|), quantum state restoration algorithm extends a large subsystem to the full state efficiently. This enables copying small subsystems and performing tomography on them in polynomial time. The approach also allows estimating statistics of any efficiently implementable POVM on |ψ⟩ in time polynomial in the POVM's number of outcomes.

Nonadditive Quantum Codes Surpass 1/2 Rate for Amplitude Damping Channel Correction

Presents a family of nonadditive quantum error-correcting codes tailored to the amplitude damping channel, leveraging self-complementarity to correct all first-order errors. These codes achieve rates exceeding 1/2, outperforming traditional additive codes in this noise model. Recovery operations are generated via a simple projection-based algorithm, facilitating practical implementation.

Nonadditive Quantum Codes Outperform Additive Ones for Amplitude Damping Error Correction

Researchers construct families of nonadditive quantum codes that correct single amplitude damping errors and outperform the best additive codes in encoded dimension. One family derives from nonlinear classical codes for asymmetric channels, yielding superior parameters for block lengths n > 7 except n=2^r-1. A GF(3)-based generalization produces even better codewords-stabilized (CWS) codes up to n=14, enabling straightforward encoding and decoding circuits.

Random Initial Hamiltonians Mitigate Small-Gap Obstacles in Quantum Adiabatic 3SAT Solving

Researchers construct 3SAT instances with two planted solutions, one penalized by a single clause, that resist efficient solving by the standard quantum adiabatic algorithm due to small spectral gaps. Randomly modifying the initial Hamiltonian yields adiabatic paths avoiding these gaps with substantial probability, as supported by Quantum Monte Carlo simulations up to 150 spins. This implies running the algorithm with multiple random paths per instance to enhance performance, though its efficacy on fully random 3SAT remains unproven.

Semidefinite Programming Optimizes Quantum Error Recovery for Maximum Entanglement Fidelity

Quantum error correction traditionally embeds states in a coded subspace and corrects a predefined error set perfectly. This work optimizes the recovery operation to maximize entanglement fidelity for any given input state and noise model, rather than perfect correction on subsets. The optimization is formulated as a semidefinite program (SDP), enabling efficient convex optimization solutions, and interprets recovery as quantum diversity combining post a spreading channel.

Generalized Concatenation Yields Optimal Nonadditive Quantum Codes Surpassing Stabilizers

Generalized concatenated quantum codes extend concatenation to both stabilizer and nonadditive codes, providing a systematic construction method. New families of single-error-correcting nonadditive codes are built for binary and nonbinary alphabets. These codes exceed stabilizer code performance for finite block lengths and achieve the quantum Hamming bound asymptotically.

Generalized Concatenation Builds Strong Quantum Error-Correcting Codes with Classical Outer Codes

Generalized concatenation constructs quantum error-correcting codes by combining quantum inner codes with linear or nonlinear classical outer codes. This method yields new high-performance codes, including stabilizer codes and nonadditive quantum codes. The approach expands beyond traditional restrictions on outer codes, enabling broader code design possibilities.

Generalized Stabilizer Codes for Amplitude Damping Channels with Optimizable Recovery

Presents channel-adapted quantum error correction using stabilizer formalism for the amplitude damping channel. Generalizes the [4,1] approximate code into a [2(M+1),M] family and introduces a [7,3] code derived from the classical Hamming code. All codes employ Clifford-group operations for encoding and recovery, with optional adaptation to damping probability gamma for enhanced performance.

Quantum Reverse Shannon Theorems Quantify Resource Tradeoffs for Noisy Channel Simulation

Quantum reverse Shannon theorems characterize efficient simulation of noisy channels using noiseless ones or other noisy channels, requiring auxiliary resources like shared randomness classically or entanglement quantumly. For tensor-power sources, ebits suffice; general sources demand additional resources such as entanglement-embezzling states or backward communication. The paper provides single-letter expressions for communication costs and resource tradeoffs, including a new formula for coherent feedback excess cost and a strong converse to the entanglement-assisted capacity theorem.

Two-Way Classical Communication Boosts Entanglement Purification and Quantum Capacity Bounds for Depolarizing Channel

Leung and Shor introduce an enhanced protocol for purifying bipartite mixed state entanglement using two-way classical communication. This protocol outperforms prior methods and establishes a tighter lower bound on the quantum capacity of the quantum depolarizing channel under two-way assistance. The advancement leverages bidirectional feedback to improve distillation efficiency in noisy quantum channels.

iTEBD Algorithm Extended to Infinite Trees Reveals Phase Transition Differences in Quantum Transverse Ising Model

Generalizes Vidal's iTEBD algorithm from infinite spin chains to infinite tree geometries using Matrix Product States for efficient simulation on Bethe lattices. Numerically observes a second-order phase transition in the transverse-field Ising model on trees, differing from the 1D chain case. Further examines a variant with longitudinal field exhibiting highly degenerate ground state when transverse field vanishes, contrasting the pure Ising model's double degeneracy.

PPT Test Fails Spectacularly as Approximation for Separable States in High Dimensions

The positive partial transpose (PPT) criterion, a standard entanglement detection test, provides no effective bound on the trace distance from separable states, with the maximum distance approaching 1 as system dimension grows to infinity. Similar limitations apply to other separability criteria including reduction, majorization, and symmetric extension tests. Evidence suggests PPT states and separable states exhibit fundamentally different geometric structures.

Jones Polynomial Approximation at Fifth Root of Unity is Complete for One Clean Qubit Model

Approximating the Jones polynomial at a fifth root of unity for the trace closure of a braid is BQP-complete for the one clean qubit model. One clean qubit computers can compute this approximation in polynomial time in braid strands and crossings. Conversely, simulating one clean qubit computation reduces to this approximation problem, establishing exact completeness.

Graph Concatenation via Generalized Local Complementation Enables Systematic Construction of Large Quantum Codes

Stabilizer quantum codes are locally equivalent to graph codes, and codeword stabilized codes are described by graphs paired with classical codes. The paper introduces graph concatenation, using generalized local complementation to combine inner and outer code graphs, providing a systematic method for constructing concatenated quantum codes. This approach supports both binary and non-binary codes, along with their generalizations, targeting good quantum codes with large block lengths.

Backward Classical Communication Separates Quantum Capacities of Erasure Channels

Leung, Lim, and Shor develop a protocol for the quantum erasure channel with backward classical communication that exceeds prior achievable rates. They prove an upper bound on this capacity, which is strictly lower than the capacity under two-way classical assistance. This establishes a separation between one-way backward and full two-way classical communication for quantum capacity.

Quantum Clique is QMA-Complete; Holevo Capacity and Minimum Entropy are NP-Complete

The quantum clique problem, defined as deciding whether a quantum channel admits k zero-error distinguishable states, is QMA-complete, providing a quantum analogue to the classical NP-complete clique problem via zero-error channel capacity. Computing the Holevo capacity and minimum entropy of quantum channels is NP-complete, with the result holding even for entanglement-breaking channels. These hardness results highlight fundamental computational barriers in quantum information theory.

Collision-Free Quantum Money: Disproving Prior Schemes and Proposing Novel Unclonable Protocols

The paper demonstrates insecurity in the sole published public-key quantum money scheme, where a bank issues quantum states verifiable by all but clonable or forgeable by others. It introduces collision-free quantum money protocols, where even the bank cannot produce multiple identical-looking money states. A blueprint and a concrete (potentially insecure) example are provided as steps toward a new quantum cryptographic protocol.

Gaussian Random Codes Prove Coherent Information as Achievable Rate for Quantum Channel Entanglement Transmission

Hayden, Shor, and Winter construct random quantum codes from Gaussian ensembles, using Haar-distributed subspaces distorted by the input density operator, to achieve the coherent information rate for entanglement transmission over noisy quantum channels. Large deviations ensure low-error decoding of classical data in Fourier-conjugate bases, while an information-uncertainty relation guarantees high quantum mutual information post-transmission. Monogamy of correlations then implies negligible channel-environment entanglement, enabling receiver decoding.

Unentangled Quantum Proofs Scale Sublinearly for SAT Verification

A verifier can confirm 3SAT satisfiability using O(sqrt(n)) unentangled quantum proofs of O(log n) qubits each, with constant soundness error, leveraging short PCPs. Under a weak Additivity Conjecture, QMA(2) protocols amplify to exponential error reduction, implying QMA(k)=QMA(2) for k>=2. Perfect disentanglers cannot simulate multiple unentangled proofs with a single entangled one.

Proof Establishes C3 Gates as Generalized Semi-Clifford Operators

Beigi and Shor prove that all C3 gates in the Gottesman-Chuang hierarchy are generalized semi-Clifford operators, confirming Zeng et al.'s conjecture for k=3. This result leverages teleportation on stabilizer codes, extending fault-tolerant application beyond Pauli and Clifford gates. The techniques provide deeper characterization of C3 gates and intuition for tackling the conjecture at higher k ≥ 4.

Adaptive Protocols Boost Entanglement Purification Yields Across Fidelities

Leung and Shor introduce a family of entanglement purification protocols generalizing four prior methods (recurrence, modified recurrence, Maneva-Smolin, Leung-Shor) using two-way classical communication. These protocols achieve higher yields than predecessors over a broad range of initial fidelities F and surpass universal hashing yields for F < 0.993 and as F approaches 1. The improved yields establish new lower bounds on the two-way assisted quantum capacity of the depolarizing channel.

Structured Algorithms Enable Scalable Near-Optimal Quantum Error Recovery Tailored to Channel Models

Numerical algorithms adapt quantum error correction to specific channel models by computing structured recovery operations that start with projective error syndrome measurements, avoiding the computational intractability of full semidefinite program (SDP) optimization for long codes. These methods produce intuitive, physically meaningful CPTP recovery maps with performance certified near-optimal via Lagrange duality bounds. They enhance scalability while preserving high average entanglement fidelity.

Adaptive Classical Capacity Bridges Separate and Joint Measurement Limits for Symmetric Quantum States

Shor defines a new adaptive classical capacity C_{1,A} for quantum channels transmitting three symmetric pure states in three dimensions. This capacity lies strictly between C_{1,1} (separate measurements per state) and C_{1,infinity} (joint measurements on all received states). The protocol uses staged measurements on individual signals, where later stages adapt based on classical computations from prior outcomes, enabling higher rates than separate measurements without full joint access.

Stabilizer Codes Enable Constant Energy Gaps for Fault-Tolerant Adiabatic Quantum Computation

Stabilizer codes provide a constant energy gap in adiabatic quantum Hamiltonians, protecting against 1-local and 2-local noise. These codes transform universal 2-local adiabatic Hamiltonians (with inverse-polynomial gaps) into fault-tolerant 4-local and 6-local Hamiltonians, respectively. This achieves optimal locality within the adiabatic framework while maintaining resistance to noise via the constant gap.

Quantum Channel Classical Capacity Trade-Off with Limited Entanglement Assistance

Peter Shor's paper derives the full trade-off curve for the classical capacity of a quantum channel as a function of the entanglement consumed by sender and receiver. The curve endpoints are the unassisted Holevo-Schumacher-Westmoreland (HSW) capacity and the entanglement-assisted capacity, defined as the maximum quantum mutual information over input density matrices. The proof leverages the HSW formula and yields a simpler derivation of the entanglement-assisted capacity formula.

Superdense Coding Boosts Capacity of Lossy Bosonic Channels Above 50% Efficiency

The paper computes the entanglement-assisted capacity of a multimode bosonic channel with loss, showing that for channel efficiencies η > 50%, superdense coding enables transmission of more bits than the number storable in the sent message. Bounds are also derived for other capacities of this multimode channel. This demonstrates quantum advantage in broadband lossy communication via pre-shared entanglement.

Concatenated Cat Codes Beat Quantum Hashing Bound for Highly Noisy Depolarizing Channels

DiVincenzo, Shor, and Smolin introduce additive quantum error-correcting codes using cat code concatenation within random hashing that achieve positive capacity for depolarizing channels at fidelity f > 0.80944, surpassing the quantum random coding threshold of f > 0.81071. A block size-5 cat code proves optimal for single concatenation, yielding the best multiple-concatenated code at block size 25. They derive a relation linking attainable capacity to the coherent information of inner code states.

Shor-Preskill Proof Establishes BB84 Security via Entanglement Purification and CSS Codes

Shor and Preskill prove BB84 quantum key distribution secure by linking it to an entanglement purification protocol whose security follows from Lo-Chau methods. The purification protocol employs CSS codes to eliminate the need for quantum computation. This establishes unconditional security for BB84 against general attacks.

Optimal Packings Achieving Orthoplex Bound in Grassmannians for Powers of 2

Shor and Sloane identify a family of optimal packings in Grassmannian manifolds, packing (m² + m - 2) subspaces of dimension m/2 into m-dimensional space when m is a power of 2. These packings achieve the orthoplex bound, establishing optimality. The discovery stems from a remarkable coincidence in combinatorial geometry.

Group-Theoretic Framework Enables New Quantum Error-Correcting Codes with Improved Parameters

Introduces a group-theoretic framework using orthogonal geometry to simplify descriptions of quantum error-correcting codes and construct new ones. Demonstrates codes encoding 3 logical qubits into 8 physical qubits correcting 1 error, 4 into 10 correcting 1 error, 1 into 13 correcting 2 errors, and 1 into 29 correcting 5 errors. Published in Phys. Rev. Lett. after revisions acknowledging slightly worse rates than initially claimed.

Unitary Gates Exhibit Dramatic Capacity Asymmetries Under Time-Reversal and Exchange Symmetries

Unitary gates in quantum communication display significant separations in capacities due to time-reversal (forward vs. backward, even with entanglement assistance) and exchange (Alice-Bob) symmetries. A general time-reversal rule relates the capacities of a unitary gate to its inverse, explaining prior failures to find asymmetries. Erasing quantum information and destroying entanglement emerges as a valuable resource for enhancing communication capacities.

Few Quantum Algorithms Outpace Classical Counterparts, with Open Questions on Rarity

Peter Shor's notes survey quantum algorithms achieving significant speedups over classical methods, noting only a handful of such techniques have been discovered. It remains unclear if this scarcity stems from limited researcher intuition or inherent limitations on problems amenable to quantum acceleration. The work serves as an introduction delivered at the 2000 AMS short course.

Additivity Proven for Classical Capacity and Minimum Output Entropy of Entanglement-Breaking Quantum Channels

Peter Shor's paper proves additivity of both the minimum output entropy and the Holevo-Schumacher-Westmoreland classical capacity for tensor products where one channel is entanglement-breaking and the other is arbitrary. It also provides a bound on subadditivity of minimum output entropy and superadditivity of channel capacity for products of two arbitrary quantum channels, using entanglement of formation. These results advance understanding of quantum channel capacities in information theory.

Optimal 1 cbit + 1 ebit per Qubit for Universal Remote State Preparation

Remote state preparation (RSP) enables sending known quantum states using less classical communication than teleportation, with a fundamental trade-off between cbits and ebts. The paper derives optimal RSP for arbitrary many-qubit states requiring exactly 1 cbit and 1 ebit per qubit in the universal setting, proven via a dichotomy showing these resource bounds are simultaneously tight. This yields the exact trade-off curve for ensembles of pure states and entangled states, building on quantum-classical trade-offs in data compression.

First Superior Nonadditive Quantum Code Encodes 6 States in 5 Qubits with Single-Qubit Erasure Correction

This paper introduces the first quantum error-correcting code that outperforms all prior stabilizer (additive) codes. The nonadditive code encodes 6 logical states into 5 physical qubits and corrects any single qubit erasure. It breaks the limitation of Abelian stabilizer formalisms using tensor products of Pauli matrices.

Shor Equates Four Major Additivity Conjectures in Quantum Information Theory

Peter Shor proves the equivalence of four key additivity conjectures in quantum information theory. These include additivity of minimum output entropy for quantum channels, additivity of the Holevo expression for classical capacity, additivity of entanglement of formation, and strong superadditivity of entanglement of formation. All four conjectures are either true or false together, reducing the independent open problems from many to one in this cluster.

Assisted Classical Side Channels Boost Quantum Channel Capacities Beyond Unassisted Holevo Limit

Discrete memoryless quantum channels are constructed where the two-way classical assisted quantum capacity Q_2 exceeds the unassisted Holevo capacity C_H, via a control input/output that enables coordinated correction of data evolution using entanglement and classical communication. The same channels show feedback-assisted classical capacity C_B and quantum capacity Q_B surpassing C_H. A related dephasing channel has unassisted classical capacity C equal to C_H but strictly less than the independent classical assisted capacity C_2.

Group-Theoretic Construction Yields Optimal Grassmannian Packings via Clifford Groups

The paper develops a group-theoretic framework using totally isotropic subspaces in the orthogonal space Ω⁺(2i,2) to construct infinite families of packings consisting of 2^k-dimensional subspaces in ℝ^{2^i}. A specific Clifford group drives these constructions, establishing optimality for select families. These packings connect to Barnes-Wall lattices, Kerdock sets, and quantum-error-correcting codes, unifying discrete geometry with coding theory.

GF(4) Additive Self-Orthogonal Codes Enable Quantum Error Correction

Quantum error-correcting codes are constructed from additive codes over GF(4) that are self-orthogonal under a trace inner product. This equivalence enables systematic discovery of new quantum codes and derivation of bounds. The paper presents multiple new codes and a comprehensive table of upper and lower bounds for codes up to 30 qubits.

Canonical NPT States in d⊗d Systems Exhibit Evidence of Bound Entanglement

DiVincenzo et al. introduce a two-parameter family of bipartite mixed states ρ_bc in d⊗d Hilbert space that are NPT but conjectured to be undistillable into 2⊗2 maximally entangled states via LOCC. For specific states in this family, no entanglement can be extracted from any number of copies using 2⊗2 projections, supporting undistillability. These states are canonical: any NPT bipartite mixed state can be reduced to ρ_bc form via LOCC, and their distillability equates to an open question on composed positive linear maps.

Bound Entanglement Superactivation: Two Non-Distillable States Yield Distillable Entanglement

In a multi-party quantum setting, tensoring two bound-entangled (non-distillable) states produces a distillable state, demonstrating genuine superadditivity of distillable entanglement. This superactivation reveals that entanglement distillation rates are not merely additive but can exceed individual contributions. Additionally, unlockable bound-entangled states are proven asymptotically non-separable, confirming their irreducible bound nature.

Halving Key Requirements for Near-Perfect Quantum State Randomization

A set of O(d log d) unitary operators suffices to nearly perfectly randomize any input pure state in dimension d, reducing from the Θ(d²) unitaries needed for perfect randomization. This halves the shared random key bits required for near-perfect private quantum channels from 2 log d. The construction enables efficient LOCC data hiding (log d qubits in 2 log d qubits), locked classical correlations with small keys, and remote state preparation with log d bits communication and log d ebits entanglement.

Oblivious Remote State Preparation Matches Teleportation's Communication Cost

For pure states whose projectors form a basis, oblivious remote state preparation protocols—using only forward communication, entanglement, deterministic exact preparation without leaking extra information—can be modified to require just one copy of the state from the sender. This modification preserves classical communication costs. Consequently, such protocols demand at least as much classical communication as teleportation, validating Lo's conjecture and doubling the cost of some non-oblivious alternatives.

Quantum MacWilliams Identities Link Fidelity Measures to Enable Code Existence Bounds

Shor and Laflamme derive identities relating entanglement fidelity and average fidelity for completely depolarizing quantum channels, establishing a quantum analog of classical MacWilliams identities. These identities connect the weight enumerator of a quantum code to its dual, mirroring classical coding theory. Linear programming techniques from classical bounds adapt to quantum settings, with demonstrated applications for investigating code existence.

Entanglement-Breaking Channels Admit Holevo Canonical Form with Additional Extreme Points Beyond CQ Maps for d>2

Entanglement-breaking channels, which map entangled states to separable ones when tensored with identity, admit a canonical Holevo form. The convex set of trace-preserving entanglement-breaking maps has classical-quantum (CQ) maps as extreme points among completely positive trace-preserving maps, but for dimensions d>2, additional non-extreme-CQ extreme points exist. The paper provides equivalent characterizations and examines special classes of these channels.

Partial Syndrome Quantum Error Correction Enables Reliable Transmission Beyond Full Syndrome Limits

Shor and Smolin introduce a quantum error-correcting code that avoids identifying the complete error syndrome, enabling reliable quantum information transmission through channels adding more than one bit of entropy per qubit. This overcomes limitations of prior codes requiring full syndrome identification. For the depolarizing channel, their code achieves reliability at fidelity 0.8096, improving on the previous best of 0.8107.

Sum-of-Squares Algorithm Achieves Near-Optimal Expected Waste for Discrete Bin Packing Distributions

The deterministic Sum of Squares (SS) algorithm for online bin packing with integral capacities runs in O(nB) time and matches the sublinear expected waste of optimal algorithms for any discrete distribution with sublinear optimal waste. For distributions with bounded optimal expected waste, SS incurs at most O(log n) expected waste. A randomized variant SS* achieves essentially optimal expected performance across all discrete distributions using an O(nB log B)-time algorithm, supported by a novel LP-based method to compute exact optimal waste growth rates.

Unextendible Product Bases Generate Bound Entangled States with Purely Multipartite Entanglement

Unextendible product bases (UPBs) are incomplete orthogonal product bases in multipartite quantum systems whose complementary subspace contains no product states. The uniform mixed state over this complementary subspace is a bound entangled state, undetectable by distillation. A specific 2x2x2 tripartite UPB yields a mixed state with genuine tripartite entanglement but separable bipartite reductions. UPB members cannot be perfectly distinguished via local POVMs and classical communication.

Quantum Channel Capacity for Joint Classical-Quantum Transmission

Devetak and Shor derive an expression for the admissible rate pairs enabling simultaneous transmission of classical and quantum information over a quantum channel, generalizing the individual classical and quantum capacities. The formula requires regularization over multiple channel uses but simplifies to a single-letter expression for generalized dephasing channels. Analogous regularized formulae are conjectured for public-private capacities and distillable common randomness with entanglement in bipartite states.

Entanglement-Assisted Classical Capacity of Quantum Channels Equals Input + Output Entropies Minus Joint Entropy

The entanglement-assisted classical capacity of a quantum channel is the maximum over input states ρ of H(ρ) + H(channel(ρ)) - H(ρ, channel(ρ)), where the joint entropy uses an entangled purification. This formula mirrors the classical channel capacity expression. The paper computes this for the qubit amplitude damping channel and the bosonic channel with amplification/attenuation and Gaussian noise, and proves that shared randomness makes all classical DMCs of equal capacity asymptotically simulable with unit efficiency.

Older entries →