kardesyazilim

Eser	Yazar/Yapımcı	Siyasi Etki
Silent Spring (1962)	Rachel Carson	Çevre bilim kurgusu olarak okunur; ABD'de DDT yasağını ve Çevre Koruma Ajansı'nın (EPA) kurulmasını tetikledi.
The Handmaid's Tale (1985)	Margaret Atwood	2017'den sonra ABD'de kadın hakları protestolarında "kırmızı elbiseli kadınlar" sembolü haline geldi; Roe v. Wade tartışmalarında referans noktası oldu.
Snow Crash (1992)	Neal Stephenson	"Metaverse" kavramını tanıttı; 2021'de Facebook'un ismini Meta olarak değiştirmesinde doğrudan etkili oldu.
The Social Dilemma (2020)	Netflix belgeseli	ABD Kongresi'nde sosyal medya düzenlemesi tartışmalarında kanıt olarak gösterildi; Avrupa Birliği'nin Dijital Hizmetler Yasası'na zemin hazırladı.

kardesyazilim

Feature	Variational Inference (VI)	Markov Chain Monte Carlo (MCMC)
Goal	Find best approximation in a tractable family	Generate exact samples from true posterior (asymptotically)
Accuracy	Biased (approximate); underestimates uncertainty	Unbiased (converges to true posterior)
Speed	Fast; scales to large datasets	Slow; often impractical for big data
Optimization	Gradient-based; deterministic	Sampling-based; stochastic
Parallelization	Easily parallelizable (e.g., mini-batches)	Hard to parallelize (chains are sequential)
Tuning	Choose variational family ( \mathcal{Q} )	Choose proposal distribution, step size, etc.
Uncertainty quantification	Can be too confident (KL(q∥p) is mode-seeking)	More reliable posterior coverage
Use cases	Real-time inference, VAEs, large-scale Bayesian models	Small-data settings, diagnostics, gold-st

kardesyazilim

Özellik	RPA	IPA
Temel Özellik	Kurala dayalı, tekrarlayan görevleri otomatikleştirir.	Yapay zeka ile desteklenmiş, karmaşık ve değişken süreçleri yönetir.
Veri Türü	Yapılandırılmış veriler (örneğin Excel, CRM veritabanları).	Yapılandırılmamış veya yarı yapılandırılmış veriler (PDF, e-posta, tarama, resim).
Esneklik	Sabit kurallara bağlıdır; değişikliklerde yeniden programlanır.	Öğrenir, uyar ve zamanla gelişir (makine öğrenimi ile).
Kullanılan Teknolojiler	Sadece RPA araçları (UIPath, Automation Anywhere, Blue Prism).	RPA + AI + ML + NLP + Bilgisayarlı Görüş + Süreç Madenciliği.
Karar Verme	Basit "eğer-ise" kurallarıyla sınırlıdır.	Tahmine dayalı analizlerle akıllı kararlar alabilir.
İşlem Karmaşıklığı	Düşük – doğrusal, basit süreçler.	Yüksek – dallanmış, dinamik, çok adımlı süreçler.

kardesyazilim

Use Case	Best Algorithm
General-purpose RL, good starting point	PPO
High-stakes environments requiring stability	PPO or TRPO
Continuous control (e.g., robotics)	SAC or DDPG
Fast prototyping or simple tasks	A2C
Importance of exploration and long-term planning	SAC
High sample efficiency required	SAC or DDPG

kardesyazilim

Feature	PPO	TRPO	DDPG	A2C (Advantage Actor-Critic)	SAC (Soft Actor-Critic)
Algorithm Type	On-policy	On-policy	Off-policy	On-policy	Off-policy
Core Idea	Clipped surrogate objective	Trust region constraint (KL divergence)	Actor-Critic + Q-learning (for continuous actions)	Synchronous advantage estimation	Maximum entropy (exploration) + off-policy
Stability	Very stable	Very stable	Can be unstable	Stable but can be sensitive to hyperparams	Very stable
Sample Efficiency	Moderate	Moderate	High (due to replay buffer)	Moderate (on-policy)	High (off-policy, replay buffer)
Complexity	Simple to implement	Complex (requires conjugate gradient)	Moderate to Complex	Complex (requires conjugate gradient)	Moderate to Complex
Action Space	Both discrete & continuous	Both discrete & continuous	Continuous only	Both discrete & continuous	Both discrete & continuous (S

kardesyazilim

Library	Strengths	Weaknesses
XGBoost	Highly customizable, GPU support, mature	Slower than LGBM on large data
LightGBM	Extremely fast, memory-efficient	Less accurate with small data
CatBoost	Best for categorical features, low tuning	Slower training, high RAM use

kardesyazilim

Method	Training Style	Error Focus	Variance	Bias	Typical Use Case
Bagging	Parallel (independent)	Reduces variance	↓↓	↔	High-variance models (e.g., deep trees)
Boosting	Sequential	Reduces bias	↓	↓↓	Weak learners; structured/tabular data
Stacking	Hybrid	Leverages diversity	↓	↓	When you have diverse strong models

---

↓ = reduction, ↔ = little change

---

kardesyazilim

#	Requirement	Verified? (Y/N)	Notes
1	HSM is FIPS 140-2 Level 3 (or FIPS 140-3 Level 3) validated		Check NIST CMVP list
2	Cryptographic keys for CHD never exist outside HSM in plaintext		Confirm via architecture review
3	All key management (generation, storage, rotation, destruction) occurs within HSM
4	HSM access is restricted via strong authentication (MFA recommended)		PCI DSS Req 8
5	Role separation enforced (e.g., SO vs. Crypto User vs. Auditor)		PCI DSS Req 7
6	All HSM operations logged; logs sent to SIEM		PCI DSS Req 10
7	HSM physically secured (if on-prem) or in compliant cloud environment		PCI DSS Req 9
8	HSM firmware/software kept up to date		PCI DSS Req 6

kardesyazilim

Criteria	Requirement	Why It Matters
FIPS 140-2/3 Validation	Must be FIPS 140-2 Level 3 (or FIPS 140-3 Level 3) validated	Required by PCI PIN and P2PE; strongly recommended for general PCI DSS key protection
Tamper Resistance	Physical and logical tamper detection/response (e.g., zeroization on breach)	Prevents key extraction if device is compromised
Secure Key Storage	Keys never leave HSM in plaintext; all crypto operations inside HSM	Meets PCI DSS Req 3.5–3.7
High Availability & Scalability	Clustering, load balancing, failover support	Ensures uptime for payment systems
APIs & Integration	Supports PKCS#11, Java JCA/JCE, Microsoft CNG, REST (for cloud)	Enables integration with apps, databases, payment switches
Audit Logging	Immutable, time-stamped logs of all operations	Supports PCI DSS Req 10 (logging & monitoring)
Role-Based Access Control (RBAC)	Separation of duties (e.g.,

kardesyazilim

Aspect	HSM	PCI DSS	ISO/IEC 27001
Nature	Technical security device	Mandatory compliance standard	Voluntary management system standard
Focus	Cryptographic key protection	Protection of cardholder data	Holistic information security
Role of HSM	Core technology	Enabler for key requirements	Risk treatment option
Certification	FIPS 140-2/3 validation	Annual assessment (SAQ or ROC)	Third-party certification (optional)